Delete docsgpt_scanner.py

-Only for api_tool for now, if this solution works well then implementation for other tools is required
- Need to check api keys creation with the current proxies
- Show connection string example at creation
- locale needs updates for other languages

.gitattributes

@@ -1,2 +0,0 @@
-# Auto detect text files and perform LF normalization
-* text=auto

.gitignore

@@ -113,7 +113,6 @@ venv.bak/
 # Spyder project settings
 .spyderproject
 .spyproject
-.jwt_secret_key
 
 # Rope project settings
 .ropeproject

README.md

@@ -48,16 +48,10 @@
 - [x] Add tools (Jan 2025)
 - [x] Manually updating chunks in the app UI (Feb 2025)
 - [x] Devcontainer for easy development (Feb 2025)
-- [x] ReACT agent (March 2025)
-- [x] Chatbots menu re-design to handle tools, agent types, and more (April 2025)
-- [x] New input box in the conversation menu (April 2025)
-- [x] Add triggerable actions / tools (webhook) (April 2025)
-- [x] Agent optimisations (May 2025)
-- [ ] Filesystem sources update (July 2025)
-- [ ] Anthropic Tool compatibility (July 2025)
-- [ ] MCP support (July 2025)
-- [ ] Add OAuth 2.0 authentication for tools and sources (August 2025)
-- [ ] Agent scheduling
+- [ ] Anthropic Tool compatibility
+- [ ] Add triggerable actions / tools (webhook)
+- [ ] Add OAuth 2.0 authentication for tools and sources
+- [ ] Chatbots menu re-design to handle tools, scheduling, and more
 
 You can find our full roadmap [here](https://github.com/orgs/arc53/projects/2). Please don't hesitate to contribute or create issues, it helps us improve DocsGPT!
 
@@ -98,15 +92,13 @@ A more detailed [Quickstart](https://docs.docsgpt.cloud/quickstart) is available
    ./setup.sh
    ```
 
+This interactive script will guide you through setting up DocsGPT. It offers four options: using the public API, running locally, connecting to a local inference engine, or using a cloud API provider.  The script will automatically configure your `.env` file and handle necessary downloads and installations based on your chosen option.
+
 **For Windows:**
 
-2. **Run the PowerShell setup script:**
+2. **Follow the Docker Deployment Guide:**
 
-   ```powershell
-   PowerShell -ExecutionPolicy Bypass -File .\setup.ps1
-   ```
-
-Either script will guide you through setting up DocsGPT. Four options available: using the public API, running locally, connecting to a local inference engine, or using a cloud API provider.  Scripts will automatically configure your `.env` file and handle necessary downloads and installations based on your chosen option.
+   Please refer to the [Docker Deployment documentation](https://docs.docsgpt.cloud/Deploying/Docker-Deploying) for detailed step-by-step instructions on setting up DocsGPT using Docker. 
 
 **Navigate to http://localhost:5173/**
 
@@ -115,7 +107,7 @@ To stop DocsGPT, open a terminal in the `DocsGPT` directory and run:
 ```bash
 docker compose -f deployment/docker-compose.yaml down
 ```
-(or use the specific `docker compose down` command shown after running the setup script).
+(or use the specific `docker compose down` command shown after running `setup.sh`).
 
 > [!Note]
 > For development environment setup instructions, please refer to the [Development Environment Guide](https://docs.docsgpt.cloud/Deploying/Development-Environment).

application/Dockerfile

@@ -84,4 +84,4 @@ EXPOSE 7091
 USER appuser
 
 # Start Gunicorn
-CMD ["gunicorn", "-w", "1", "--timeout", "120", "--bind", "0.0.0.0:7091", "--preload", "application.wsgi:app"]
+CMD ["gunicorn", "-w", "2", "--timeout", "120", "--bind", "0.0.0.0:7091", "application.wsgi:app"]

application/agents/agent_creator.py

@@ -1,11 +1,9 @@
 from application.agents.classic_agent import ClassicAgent
-from application.agents.react_agent import ReActAgent
 
 
 class AgentCreator:
     agents = {
         "classic": ClassicAgent,
-        "react": ReActAgent,
     }
 
     @classmethod
@@ -13,4 +11,7 @@ class AgentCreator:
         agent_class = cls.agents.get(type.lower())
         if not agent_class:
             raise ValueError(f"No agent class found for type {type}")
+        config = kwargs.pop('config', None)
+        if isinstance(config, dict) and 'proxy_id' in config and 'proxy_id' not in kwargs:
+            kwargs['proxy_id'] = config['proxy_id']
         return agent_class(*args, **kwargs)

application/agents/base.py

@@ -1,98 +1,54 @@
-import uuid
-from abc import ABC, abstractmethod
-from typing import Dict, Generator, List, Optional
-
-from bson.objectid import ObjectId
+from typing import Dict, Generator
 
+from application.agents.llm_handler import get_llm_handler
 from application.agents.tools.tool_action_parser import ToolActionParser
 from application.agents.tools.tool_manager import ToolManager
 
 from application.core.mongo_db import MongoDB
-from application.core.settings import settings
-
-from application.llm.handlers.handler_creator import LLMHandlerCreator
 from application.llm.llm_creator import LLMCreator
-from application.logging import build_stack_data, log_activity, LogContext
-from application.retriever.base import BaseRetriever
 
 
-class BaseAgent(ABC):
+class BaseAgent:
     def __init__(
         self,
-        endpoint: str,
-        llm_name: str,
-        gpt_model: str,
-        api_key: str,
-        user_api_key: Optional[str] = None,
-        prompt: str = "",
-        chat_history: Optional[List[Dict]] = None,
-        decoded_token: Optional[Dict] = None,
-        attachments: Optional[List[Dict]] = None,
+        endpoint,
+        llm_name,
+        gpt_model,
+        api_key,
+        user_api_key=None,
+        decoded_token=None,
+        proxy_id=None,
     ):
         self.endpoint = endpoint
-        self.llm_name = llm_name
-        self.gpt_model = gpt_model
-        self.api_key = api_key
-        self.user_api_key = user_api_key
-        self.prompt = prompt
-        self.decoded_token = decoded_token or {}
-        self.user: str = decoded_token.get("sub")
-        self.tool_config: Dict = {}
-        self.tools: List[Dict] = []
-        self.tool_calls: List[Dict] = []
-        self.chat_history: List[Dict] = chat_history if chat_history is not None else []
         self.llm = LLMCreator.create_llm(
             llm_name,
             api_key=api_key,
             user_api_key=user_api_key,
             decoded_token=decoded_token,
         )
-        self.llm_handler = LLMHandlerCreator.create_handler(
-            llm_name if llm_name else "default"
-        )
-        self.attachments = attachments or []
+        self.llm_handler = get_llm_handler(llm_name)
+        self.gpt_model = gpt_model
+        self.tools = []
+        self.tool_config = {}
+        self.tool_calls = []
+        self.proxy_id = proxy_id
 
-    @log_activity()
-    def gen(
-        self, query: str, retriever: BaseRetriever, log_context: LogContext = None
-    ) -> Generator[Dict, None, None]:
-        yield from self._gen_inner(query, retriever, log_context)
-
-    @abstractmethod
-    def _gen_inner(
-        self, query: str, retriever: BaseRetriever, log_context: LogContext
-    ) -> Generator[Dict, None, None]:
-        pass
-
-    def _get_tools(self, api_key: str = None) -> Dict[str, Dict]:
-        mongo = MongoDB.get_client()
-        db = mongo[settings.MONGO_DB_NAME]
-        agents_collection = db["agents"]
-        tools_collection = db["user_tools"]
-
-        agent_data = agents_collection.find_one({"key": api_key or self.user_api_key})
-        tool_ids = agent_data.get("tools", []) if agent_data else []
-
-        tools = (
-            tools_collection.find(
-                {"_id": {"$in": [ObjectId(tool_id) for tool_id in tool_ids]}}
-            )
-            if tool_ids
-            else []
-        )
-        tools = list(tools)
-        tools_by_id = {str(tool["_id"]): tool for tool in tools} if tools else {}
-
-        return tools_by_id
+    def gen(self, *args, **kwargs) -> Generator[Dict, None, None]:
+        raise NotImplementedError('Method "gen" must be implemented in the child class')
 
     def _get_user_tools(self, user="local"):
         mongo = MongoDB.get_client()
-        db = mongo[settings.MONGO_DB_NAME]
+        db = mongo["docsgpt"]
         user_tools_collection = db["user_tools"]
         user_tools = user_tools_collection.find({"user": user, "status": True})
         user_tools = list(user_tools)
-
-        return {str(i): tool for i, tool in enumerate(user_tools)}
+        tools_by_id = {str(tool["_id"]): tool for tool in user_tools}
+        if hasattr(self, 'proxy_id') and self.proxy_id:
+            for tool_id, tool in tools_by_id.items():
+                if 'config' not in tool:
+                    tool['config'] = {}
+                tool['config']['proxy_id'] = self.proxy_id
+        return tools_by_id
 
     def _build_tool_parameters(self, action):
         params = {"type": "object", "properties": {}, "required": []}
@@ -136,15 +92,6 @@ class BaseAgent(ABC):
         parser = ToolActionParser(self.llm.__class__.__name__)
         tool_id, action_name, call_args = parser.parse_args(call)
 
-        call_id = getattr(call, "id", None) or str(uuid.uuid4())
-        tool_call_data = {
-            "tool_name": tools_dict[tool_id]["name"],
-            "call_id": call_id,
-            "action_name": f"{action_name}_{tool_id}",
-            "arguments": call_args,
-        }
-        yield {"type": "tool_call", "data": {**tool_call_data, "status": "pending"}}
-
         tool_data = tools_dict[tool_id]
         action_data = (
             tool_data["config"]["actions"][action_name]
@@ -169,12 +116,14 @@ class BaseAgent(ABC):
                 for param, details in action_data[param_type]["properties"].items():
                     if param not in call_args and "value" in details:
                         target_dict[param] = details["value"]
+
         for param, value in call_args.items():
             for param_type, target_dict in param_types.items():
                 if param_type in action_data and param in action_data[param_type].get(
                     "properties", {}
                 ):
                     target_dict[param] = value
+
         tm = ToolManager(config={})
         tool = tm.load_tool(
             tool_data["name"],
@@ -184,6 +133,7 @@ class BaseAgent(ABC):
                     "method": tool_data["config"]["actions"][action_name]["method"],
                     "headers": headers,
                     "query_params": query_params,
+                    "proxy_id": self.proxy_id,
                 }
                 if tool_data["name"] == "api_tool"
                 else tool_data["config"]
@@ -197,130 +147,15 @@ class BaseAgent(ABC):
         else:
             print(f"Executing tool: {action_name} with args: {call_args}")
             result = tool.execute_action(action_name, **parameters)
-        tool_call_data["result"] = (
-            f"{str(result)[:50]}..." if len(str(result)) > 50 else result
-        )
+        call_id = getattr(call, "id", None)
 
-        yield {"type": "tool_call", "data": {**tool_call_data, "status": "completed"}}
+        tool_call_data = {
+            "tool_name": tool_data["name"],
+            "call_id": call_id if call_id is not None else "None",
+            "action_name": f"{action_name}_{tool_id}",
+            "arguments": call_args,
+            "result": result,
+        }
         self.tool_calls.append(tool_call_data)
 
         return result, call_id
-
-    def _get_truncated_tool_calls(self):
-        return [
-            {
-                **tool_call,
-                "result": (
-                    f"{str(tool_call['result'])[:50]}..."
-                    if len(str(tool_call["result"])) > 50
-                    else tool_call["result"]
-                ),
-                "status": "completed",
-            }
-            for tool_call in self.tool_calls
-        ]
-
-    def _build_messages(
-        self,
-        system_prompt: str,
-        query: str,
-        retrieved_data: List[Dict],
-    ) -> List[Dict]:
-        docs_together = "\n".join([doc["text"] for doc in retrieved_data])
-        p_chat_combine = system_prompt.replace("{summaries}", docs_together)
-        messages_combine = [{"role": "system", "content": p_chat_combine}]
-
-        for i in self.chat_history:
-            if "prompt" in i and "response" in i:
-                messages_combine.append({"role": "user", "content": i["prompt"]})
-                messages_combine.append({"role": "assistant", "content": i["response"]})
-            if "tool_calls" in i:
-                for tool_call in i["tool_calls"]:
-                    call_id = tool_call.get("call_id") or str(uuid.uuid4())
-
-                    function_call_dict = {
-                        "function_call": {
-                            "name": tool_call.get("action_name"),
-                            "args": tool_call.get("arguments"),
-                            "call_id": call_id,
-                        }
-                    }
-                    function_response_dict = {
-                        "function_response": {
-                            "name": tool_call.get("action_name"),
-                            "response": {"result": tool_call.get("result")},
-                            "call_id": call_id,
-                        }
-                    }
-
-                    messages_combine.append(
-                        {"role": "assistant", "content": [function_call_dict]}
-                    )
-                    messages_combine.append(
-                        {"role": "tool", "content": [function_response_dict]}
-                    )
-        messages_combine.append({"role": "user", "content": query})
-        return messages_combine
-
-    def _retriever_search(
-        self,
-        retriever: BaseRetriever,
-        query: str,
-        log_context: Optional[LogContext] = None,
-    ) -> List[Dict]:
-        retrieved_data = retriever.search(query)
-        if log_context:
-            data = build_stack_data(retriever, exclude_attributes=["llm"])
-            log_context.stacks.append({"component": "retriever", "data": data})
-        return retrieved_data
-
-    def _llm_gen(self, messages: List[Dict], log_context: Optional[LogContext] = None):
-        gen_kwargs = {"model": self.gpt_model, "messages": messages}
-
-        if (
-            hasattr(self.llm, "_supports_tools")
-            and self.llm._supports_tools
-            and self.tools
-        ):
-            gen_kwargs["tools"] = self.tools
-        resp = self.llm.gen_stream(**gen_kwargs)
-
-        if log_context:
-            data = build_stack_data(self.llm, exclude_attributes=["client"])
-            log_context.stacks.append({"component": "llm", "data": data})
-        return resp
-
-    def _llm_handler(
-        self,
-        resp,
-        tools_dict: Dict,
-        messages: List[Dict],
-        log_context: Optional[LogContext] = None,
-        attachments: Optional[List[Dict]] = None,
-    ):
-        resp = self.llm_handler.process_message_flow(
-            self, resp, tools_dict, messages, attachments, True
-        )
-        if log_context:
-            data = build_stack_data(self.llm_handler, exclude_attributes=["tool_calls"])
-            log_context.stacks.append({"component": "llm_handler", "data": data})
-        return resp
-
-    def _handle_response(self, response, tools_dict, messages, log_context):
-        if isinstance(response, str):
-            yield {"answer": response}
-            return
-        if hasattr(response, "message") and getattr(response.message, "content", None):
-            yield {"answer": response.message.content}
-            return
-        processed_response_gen = self._llm_handler(
-            response, tools_dict, messages, log_context, self.attachments
-        )
-
-        for event in processed_response_gen:
-            if isinstance(event, str):
-                yield {"answer": event}
-            elif hasattr(event, "message") and getattr(event.message, "content", None):
-                yield {"answer": event.message.content}
-            elif isinstance(event, dict) and "type" in event:
-                yield event

application/agents/classic_agent.py

@@ -1,53 +1,141 @@
+import uuid
 from typing import Dict, Generator
-from application.agents.base import BaseAgent
-from application.logging import LogContext
-from application.retriever.base import BaseRetriever
-import logging
 
-logger = logging.getLogger(__name__)
+from application.agents.base import BaseAgent
+from application.logging import build_stack_data, log_activity, LogContext
+
+from application.retriever.base import BaseRetriever
 
 
 class ClassicAgent(BaseAgent):
-    """A simplified agent with clear execution flow.
+    def __init__(
+        self,
+        endpoint,
+        llm_name,
+        gpt_model,
+        api_key,
+        user_api_key=None,
+        prompt="",
+        chat_history=None,
+        decoded_token=None,
+        proxy_id=None,
+    ):
+        super().__init__(
+            endpoint, llm_name, gpt_model, api_key, user_api_key, decoded_token, proxy_id
+        )
+        self.user = decoded_token.get("sub")
+        self.prompt = prompt
+        self.chat_history = chat_history if chat_history is not None else []
 
-    Usage:
-    1. Processes a query through retrieval
-    2. Sets up available tools
-    3. Generates responses using LLM
-    4. Handles tool interactions if needed
-    5. Returns standardized outputs
-
-    Easy to extend by overriding specific steps.
-    """
+    @log_activity()
+    def gen(
+        self, query: str, retriever: BaseRetriever, log_context: LogContext = None
+    ) -> Generator[Dict, None, None]:
+        yield from self._gen_inner(query, retriever, log_context)
 
     def _gen_inner(
         self, query: str, retriever: BaseRetriever, log_context: LogContext
     ) -> Generator[Dict, None, None]:
-        # Step 1: Retrieve relevant data
         retrieved_data = self._retriever_search(retriever, query, log_context)
 
-        # Step 2: Prepare tools
-        tools_dict = (
-            self._get_user_tools(self.user)
-            if not self.user_api_key
-            else self._get_tools(self.user_api_key)
-        )
+        docs_together = "\n".join([doc["text"] for doc in retrieved_data])
+        p_chat_combine = self.prompt.replace("{summaries}", docs_together)
+        messages_combine = [{"role": "system", "content": p_chat_combine}]
+
+        if len(self.chat_history) > 0:
+            for i in self.chat_history:
+                if "prompt" in i and "response" in i:
+                    messages_combine.append({"role": "user", "content": i["prompt"]})
+                    messages_combine.append(
+                        {"role": "assistant", "content": i["response"]}
+                    )
+                if "tool_calls" in i:
+                    for tool_call in i["tool_calls"]:
+                        call_id = tool_call.get("call_id")
+                        if call_id is None or call_id == "None":
+                            call_id = str(uuid.uuid4())
+
+                        function_call_dict = {
+                            "function_call": {
+                                "name": tool_call.get("action_name"),
+                                "args": tool_call.get("arguments"),
+                                "call_id": call_id,
+                            }
+                        }
+                        function_response_dict = {
+                            "function_response": {
+                                "name": tool_call.get("action_name"),
+                                "response": {"result": tool_call.get("result")},
+                                "call_id": call_id,
+                            }
+                        }
+
+                        messages_combine.append(
+                            {"role": "assistant", "content": [function_call_dict]}
+                        )
+                        messages_combine.append(
+                            {"role": "tool", "content": [function_response_dict]}
+                        )
+        messages_combine.append({"role": "user", "content": query})
+
+        tools_dict = self._get_user_tools(self.user)
         self._prepare_tools(tools_dict)
 
-        # Step 3: Build and process messages
-        messages = self._build_messages(self.prompt, query, retrieved_data)
-        llm_response = self._llm_gen(messages, log_context)
+        resp = self._llm_gen(messages_combine, log_context)
 
-        # Step 4: Handle the response
-        yield from self._handle_response(
-            llm_response, tools_dict, messages, log_context
-        )
+        if isinstance(resp, str):
+            yield {"answer": resp}
+            return
+        if (
+            hasattr(resp, "message")
+            and hasattr(resp.message, "content")
+            and resp.message.content is not None
+        ):
+            yield {"answer": resp.message.content}
+            return
+
+        resp = self._llm_handler(resp, tools_dict, messages_combine, log_context)
+
+        if isinstance(resp, str):
+            yield {"answer": resp}
+        elif (
+            hasattr(resp, "message")
+            and hasattr(resp.message, "content")
+            and resp.message.content is not None
+        ):
+            yield {"answer": resp.message.content}
+        else:
+            completion = self.llm.gen_stream(
+                model=self.gpt_model, messages=messages_combine, tools=self.tools
+            )
+            for line in completion:
+                if isinstance(line, str):
+                    yield {"answer": line}
 
-        # Step 5: Return metadata
         yield {"sources": retrieved_data}
-        yield {"tool_calls": self._get_truncated_tool_calls()}
+        yield {"tool_calls": self.tool_calls.copy()}
 
-        # Log tool calls for debugging
-        log_context.stacks.append(
-            {"component": "agent", "data": {"tool_calls": self.tool_calls.copy()}}
+    def _retriever_search(self, retriever, query, log_context):
+        retrieved_data = retriever.search(query)
+        if log_context:
+            data = build_stack_data(retriever, exclude_attributes=["llm"])
+            log_context.stacks.append({"component": "retriever", "data": data})
+        return retrieved_data
+
+    def _llm_gen(self, messages_combine, log_context):
+        resp = self.llm.gen_stream(
+            model=self.gpt_model, messages=messages_combine, tools=self.tools
         )
+        if log_context:
+            data = build_stack_data(self.llm)
+            log_context.stacks.append({"component": "llm", "data": data})
+        return resp
+
+    def _llm_handler(self, resp, tools_dict, messages_combine, log_context):
+        resp = self.llm_handler.handle_response(
+            self, resp, tools_dict, messages_combine
+        )
+        if log_context:
+            data = build_stack_data(self.llm_handler)
+            log_context.stacks.append({"component": "llm_handler", "data": data})
+        return resp

application/agents/llm_handler.py

@@ -0,0 +1,254 @@
+import json
+from abc import ABC, abstractmethod
+
+from application.logging import build_stack_data
+
+
+class LLMHandler(ABC):
+    def __init__(self):
+        self.llm_calls = []
+        self.tool_calls = []
+
+    @abstractmethod
+    def handle_response(self, agent, resp, tools_dict, messages, **kwargs):
+        pass
+
+
+class OpenAILLMHandler(LLMHandler):
+    def handle_response(self, agent, resp, tools_dict, messages, stream: bool = True):
+        if not stream:
+            while hasattr(resp, "finish_reason") and resp.finish_reason == "tool_calls":
+                message = json.loads(resp.model_dump_json())["message"]
+                keys_to_remove = {"audio", "function_call", "refusal"}
+                filtered_data = {
+                    k: v for k, v in message.items() if k not in keys_to_remove
+                }
+                messages.append(filtered_data)
+
+                tool_calls = resp.message.tool_calls
+                for call in tool_calls:
+                    try:
+                        self.tool_calls.append(call)
+                        tool_response, call_id = agent._execute_tool_action(
+                            tools_dict, call
+                        )
+                        function_call_dict = {
+                            "function_call": {
+                                "name": call.function.name,
+                                "args": call.function.arguments,
+                                "call_id": call_id,
+                            }
+                        }
+                        function_response_dict = {
+                            "function_response": {
+                                "name": call.function.name,
+                                "response": {"result": tool_response},
+                                "call_id": call_id,
+                            }
+                        }
+
+                        messages.append(
+                            {"role": "assistant", "content": [function_call_dict]}
+                        )
+                        messages.append(
+                            {"role": "tool", "content": [function_response_dict]}
+                        )
+
+                    except Exception as e:
+                        messages.append(
+                            {
+                                "role": "tool",
+                                "content": f"Error executing tool: {str(e)}",
+                                "tool_call_id": call_id,
+                            }
+                        )
+                resp = agent.llm.gen_stream(
+                    model=agent.gpt_model, messages=messages, tools=agent.tools
+                )
+                self.llm_calls.append(build_stack_data(agent.llm))
+            return resp
+
+        else:
+            while True:
+                tool_calls = {}
+                for chunk in resp:
+                    if isinstance(chunk, str) and len(chunk) > 0:
+                        return
+                    elif hasattr(chunk, "delta"): 
+                        chunk_delta = chunk.delta
+
+                        if (
+                            hasattr(chunk_delta, "tool_calls")
+                            and chunk_delta.tool_calls is not None
+                        ):
+                            for tool_call in chunk_delta.tool_calls:
+                                index = tool_call.index
+                                if index not in tool_calls:
+                                    tool_calls[index] = {
+                                        "id": "",
+                                        "function": {"name": "", "arguments": ""},
+                                    }
+
+                                current = tool_calls[index]
+                                if tool_call.id:
+                                    current["id"] = tool_call.id
+                                if tool_call.function.name:
+                                    current["function"][
+                                        "name"
+                                    ] = tool_call.function.name
+                                if tool_call.function.arguments:
+                                    current["function"][
+                                        "arguments"
+                                    ] += tool_call.function.arguments
+                                tool_calls[index] = current
+
+                        if (
+                            hasattr(chunk, "finish_reason")
+                            and chunk.finish_reason == "tool_calls"
+                        ):
+                            for index in sorted(tool_calls.keys()):
+                                call = tool_calls[index]
+                                try:
+                                    self.tool_calls.append(call)
+                                    tool_response, call_id = agent._execute_tool_action(
+                                        tools_dict, call
+                                    )
+                                    if isinstance(call["function"]["arguments"], str):
+                                        call["function"]["arguments"] = json.loads(call["function"]["arguments"])
+
+                                    function_call_dict = {
+                                        "function_call": {
+                                            "name": call["function"]["name"],
+                                            "args": call["function"]["arguments"],
+                                            "call_id": call["id"],
+                                        }
+                                    }
+                                    function_response_dict = {
+                                        "function_response": {
+                                            "name": call["function"]["name"],
+                                            "response": {"result": tool_response},
+                                            "call_id": call["id"],
+                                        }
+                                    }
+
+                                    messages.append(
+                                        {
+                                            "role": "assistant",
+                                            "content": [function_call_dict],
+                                        }
+                                    )
+                                    messages.append(
+                                        {
+                                            "role": "tool",
+                                            "content": [function_response_dict],
+                                        }
+                                    )
+
+                                except Exception as e:
+                                    messages.append(
+                                        {
+                                            "role": "assistant",
+                                            "content": f"Error executing tool: {str(e)}",
+                                        }
+                                    )
+                            tool_calls = {}
+
+                        if (
+                            hasattr(chunk, "finish_reason")
+                            and chunk.finish_reason == "stop"
+                        ):
+                            return
+                    elif isinstance(chunk, str) and len(chunk) == 0:
+                            continue
+
+                resp = agent.llm.gen_stream(
+                    model=agent.gpt_model, messages=messages, tools=agent.tools
+                )
+                self.llm_calls.append(build_stack_data(agent.llm))
+
+
+class GoogleLLMHandler(LLMHandler):
+    def handle_response(self, agent, resp, tools_dict, messages, stream: bool = True):
+        from google.genai import types
+
+        while True:
+            if not stream:
+                response = agent.llm.gen(
+                    model=agent.gpt_model, messages=messages, tools=agent.tools
+                )
+                self.llm_calls.append(build_stack_data(agent.llm))
+                if response.candidates and response.candidates[0].content.parts:
+                    tool_call_found = False
+                    for part in response.candidates[0].content.parts:
+                        if part.function_call:
+                            tool_call_found = True
+                            self.tool_calls.append(part.function_call)
+                            tool_response, call_id = agent._execute_tool_action(
+                                tools_dict, part.function_call
+                            )
+                            function_response_part = types.Part.from_function_response(
+                                name=part.function_call.name,
+                                response={"result": tool_response},
+                            )
+
+                            messages.append(
+                                {"role": "model", "content": [part.to_json_dict()]}
+                            )
+                            messages.append(
+                                {
+                                    "role": "tool",
+                                    "content": [function_response_part.to_json_dict()],
+                                }
+                            )
+
+                    if (
+                        not tool_call_found
+                        and response.candidates[0].content.parts
+                        and response.candidates[0].content.parts[0].text
+                    ):
+                        return response.candidates[0].content.parts[0].text
+                    elif not tool_call_found:
+                        return response.candidates[0].content.parts
+
+                else:
+                    return response
+
+            else:
+                response = agent.llm.gen_stream(
+                    model=agent.gpt_model, messages=messages, tools=agent.tools
+                )
+                self.llm_calls.append(build_stack_data(agent.llm))
+
+                tool_call_found = False
+                for result in response:
+                    if hasattr(result, "function_call"):
+                        tool_call_found = True
+                        self.tool_calls.append(result.function_call)
+                        tool_response, call_id = agent._execute_tool_action(
+                            tools_dict, result.function_call
+                        )
+                        function_response_part = types.Part.from_function_response(
+                            name=result.function_call.name,
+                            response={"result": tool_response},
+                        )
+
+                        messages.append(
+                            {"role": "model", "content": [result.to_json_dict()]}
+                        )
+                        messages.append(
+                            {
+                                "role": "tool",
+                                "content": [function_response_part.to_json_dict()],
+                            }
+                        )
+
+                if not tool_call_found:
+                    return response
+
+
+def get_llm_handler(llm_type):
+    handlers = {
+        "openai": OpenAILLMHandler(),
+        "google": GoogleLLMHandler(),
+    }
+    return handlers.get(llm_type, OpenAILLMHandler())

application/agents/react_agent.py

@@ -1,229 +0,0 @@
-import os
-from typing import Dict, Generator, List, Any
-import logging
-
-from application.agents.base import BaseAgent
-from application.logging import build_stack_data, LogContext
-from application.retriever.base import BaseRetriever
-
-logger = logging.getLogger(__name__)
-
-current_dir = os.path.dirname(
-    os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
-)
-with open(
-    os.path.join(current_dir, "application/prompts", "react_planning_prompt.txt"), "r"
-) as f:
-    planning_prompt_template = f.read()
-with open(
-    os.path.join(current_dir, "application/prompts", "react_final_prompt.txt"),
-    "r",
-) as f:
-    final_prompt_template = f.read()
-    
-MAX_ITERATIONS_REASONING = 10
-
-class ReActAgent(BaseAgent):
-    def __init__(self, *args, **kwargs):
-        super().__init__(*args, **kwargs)
-        self.plan: str = ""
-        self.observations: List[str] = []
-
-    def _extract_content_from_llm_response(self, resp: Any) -> str:
-        """
-        Helper to extract string content from various LLM response types.
-        Handles strings, message objects (OpenAI-like), and streams.
-        Adapt stream handling for your specific LLM client if not OpenAI.
-        """
-        collected_content = []
-        if isinstance(resp, str):
-            collected_content.append(resp)
-        elif ( # OpenAI non-streaming or Anthropic non-streaming (older SDK style)
-            hasattr(resp, "message")
-            and hasattr(resp.message, "content")
-            and resp.message.content is not None
-        ):
-            collected_content.append(resp.message.content)
-        elif ( # OpenAI non-streaming (Pydantic model), Anthropic new SDK non-streaming
-            hasattr(resp, "choices") and resp.choices and
-            hasattr(resp.choices[0], "message") and
-            hasattr(resp.choices[0].message, "content") and
-            resp.choices[0].message.content is not None
-        ):
-            collected_content.append(resp.choices[0].message.content) # OpenAI
-        elif ( # Anthropic new SDK non-streaming content block
-             hasattr(resp, "content") and isinstance(resp.content, list) and resp.content and
-             hasattr(resp.content[0], "text")
-        ):
-            collected_content.append(resp.content[0].text) # Anthropic
-        else:
-            # Assume resp is a stream if not a recognized object
-            try:
-                for chunk in resp: # This will fail if resp is not iterable (e.g. a non-streaming response object)
-                    content_piece = ""
-                    # OpenAI-like stream
-                    if hasattr(chunk, 'choices') and len(chunk.choices) > 0 and \
-                       hasattr(chunk.choices[0], 'delta') and \
-                       hasattr(chunk.choices[0].delta, 'content') and \
-                       chunk.choices[0].delta.content is not None:
-                        content_piece = chunk.choices[0].delta.content
-                    # Anthropic-like stream (ContentBlockDelta)
-                    elif hasattr(chunk, 'type') and chunk.type == 'content_block_delta' and \
-                         hasattr(chunk, 'delta') and hasattr(chunk.delta, 'text'):
-                        content_piece = chunk.delta.text
-                    elif isinstance(chunk, str): # Simplest case: stream of strings
-                        content_piece = chunk
-
-                    if content_piece:
-                        collected_content.append(content_piece)
-            except TypeError: # If resp is not iterable (e.g. a final response object that wasn't caught above)
-                logger.debug(f"Response type {type(resp)} could not be iterated as a stream. It might be a non-streaming object not handled by specific checks.")
-            except Exception as e:
-                logger.error(f"Error processing potential stream chunk: {e}, chunk was: {getattr(chunk, '__dict__', chunk)}")
-
-
-        return "".join(collected_content)
-
-    def _gen_inner(
-        self, query: str, retriever: BaseRetriever, log_context: LogContext
-    ) -> Generator[Dict, None, None]:
-        # Reset state for this generation call
-        self.plan = ""
-        self.observations = []
-        retrieved_data = self._retriever_search(retriever, query, log_context)
-
-        if self.user_api_key:
-            tools_dict = self._get_tools(self.user_api_key)
-        else:
-            tools_dict = self._get_user_tools(self.user)
-        self._prepare_tools(tools_dict)
-
-        docs_together = "\n".join([doc["text"] for doc in retrieved_data])
-        iterating_reasoning = 0
-        while iterating_reasoning < MAX_ITERATIONS_REASONING:
-            iterating_reasoning += 1
-            # 1. Create Plan
-            logger.info("ReActAgent: Creating plan...")
-            plan_stream = self._create_plan(query, docs_together, log_context)
-            current_plan_parts = []
-            yield {"thought": f"Reasoning... (iteration {iterating_reasoning})\n\n"}
-            for line_chunk in plan_stream:
-                current_plan_parts.append(line_chunk)
-                yield {"thought": line_chunk}
-            self.plan = "".join(current_plan_parts)
-            if self.plan:
-                self.observations.append(f"Plan: {self.plan} Iteration: {iterating_reasoning}")
-
-
-            max_obs_len = 20000
-            obs_str = "\n".join(self.observations)
-            if len(obs_str) > max_obs_len:
-                obs_str = obs_str[:max_obs_len] + "\n...[observations truncated]"
-            execution_prompt_str = (
-                (self.prompt or "")
-                + f"\n\nFollow this plan:\n{self.plan}"
-                + f"\n\nObservations:\n{obs_str}"
-                + f"\n\nIf there is enough data to complete user query '{query}', Respond with 'SATISFIED' only. Otherwise, continue. Dont Menstion 'SATISFIED' in your response if you are not ready. "
-            )
-            
-            messages = self._build_messages(execution_prompt_str, query, retrieved_data)
-
-            resp_from_llm_gen = self._llm_gen(messages, log_context)
-
-            initial_llm_thought_content = self._extract_content_from_llm_response(resp_from_llm_gen)
-            if initial_llm_thought_content:
-                self.observations.append(f"Initial thought/response: {initial_llm_thought_content}")
-            else:
-                logger.info("ReActAgent: Initial LLM response (before handler) had no textual content (might be only tool calls).")
-            resp_after_handler = self._llm_handler(resp_from_llm_gen, tools_dict, messages, log_context)
-            
-            for tool_call_info in self.tool_calls: # Iterate over self.tool_calls populated by _llm_handler
-                observation_string = (
-                    f"Executed Action: Tool '{tool_call_info.get('tool_name', 'N/A')}' "
-                    f"with arguments '{tool_call_info.get('arguments', '{}')}'. Result: '{str(tool_call_info.get('result', ''))[:200]}...'"
-                )
-                self.observations.append(observation_string)
-
-            content_after_handler = self._extract_content_from_llm_response(resp_after_handler)
-            if content_after_handler:
-                self.observations.append(f"Response after tool execution: {content_after_handler}")
-            else:
-                logger.info("ReActAgent: LLM response after handler had no textual content.")
-
-            if log_context:
-                log_context.stacks.append(
-                    {"component": "agent_tool_calls", "data": {"tool_calls": self.tool_calls.copy()}}
-                )
-
-            yield {"sources": retrieved_data}
-
-            display_tool_calls = []
-            for tc in self.tool_calls:
-                cleaned_tc = tc.copy()
-                if len(str(cleaned_tc.get("result", ""))) > 50:
-                    cleaned_tc["result"] = str(cleaned_tc["result"])[:50] + "..."
-                display_tool_calls.append(cleaned_tc)
-            if display_tool_calls:
-                yield {"tool_calls": display_tool_calls}
-            
-            if "SATISFIED" in content_after_handler:
-                logger.info("ReActAgent: LLM satisfied with the plan and data. Stopping reasoning.")
-                break
-
-        # 3. Create Final Answer based on all observations
-        final_answer_stream = self._create_final_answer(query, self.observations, log_context)
-        for answer_chunk in final_answer_stream:
-            yield {"answer": answer_chunk}
-        logger.info("ReActAgent: Finished generating final answer.")
-
-    def _create_plan(
-        self, query: str, docs_data: str, log_context: LogContext = None
-    ) -> Generator[str, None, None]:
-        plan_prompt_filled = planning_prompt_template.replace("{query}", query)
-        if "{summaries}" in plan_prompt_filled:
-            summaries = docs_data if docs_data else "No documents retrieved."
-            plan_prompt_filled = plan_prompt_filled.replace("{summaries}", summaries)
-        plan_prompt_filled = plan_prompt_filled.replace("{prompt}", self.prompt or "")
-        plan_prompt_filled = plan_prompt_filled.replace("{observations}", "\n".join(self.observations))
-
-        messages = [{"role": "user", "content": plan_prompt_filled}]
-
-        plan_stream_from_llm = self.llm.gen_stream(
-            model=self.gpt_model, messages=messages, tools=getattr(self, 'tools', None) # Use self.tools
-        )
-        if log_context:
-            data = build_stack_data(self.llm)
-            log_context.stacks.append({"component": "planning_llm", "data": data})
-
-        for chunk in plan_stream_from_llm:
-            content_piece = self._extract_content_from_llm_response(chunk)
-            if content_piece:
-                yield content_piece
-
-    def _create_final_answer(
-        self, query: str, observations: List[str], log_context: LogContext = None
-    ) -> Generator[str, None, None]:
-        observation_string = "\n".join(observations)
-        max_obs_len = 10000
-        if len(observation_string) > max_obs_len:
-            observation_string = observation_string[:max_obs_len] + "\n...[observations truncated]"
-            logger.warning("ReActAgent: Truncated observations for final answer prompt due to length.")
-
-        final_answer_prompt_filled = final_prompt_template.format(
-            query=query, observations=observation_string
-        )
-
-        messages = [{"role": "user", "content": final_answer_prompt_filled}]
-
-        # Final answer should synthesize, not call tools.
-        final_answer_stream_from_llm = self.llm.gen_stream(
-            model=self.gpt_model, messages=messages, tools=None
-        )
-        if log_context:
-            data = build_stack_data(self.llm)
-            log_context.stacks.append({"component": "final_answer_llm", "data": data})
-
-        for chunk in final_answer_stream_from_llm:
-            content_piece = self._extract_content_from_llm_response(chunk)
-            if content_piece:
-                yield content_piece

application/agents/tools/api_tool.py

@@ -23,15 +23,43 @@ class APITool(Tool):
         )
 
     def _make_api_call(self, url, method, headers, query_params, body):
+        sanitized_headers = {}
+        for key, value in headers.items():
+            if isinstance(value, str):
+                sanitized_value = value.encode('latin-1', errors='ignore').decode('latin-1')
+                sanitized_headers[key] = sanitized_value
+            else:
+                sanitized_headers[key] = value
+
         if query_params:
             url = f"{url}?{requests.compat.urlencode(query_params)}"
-        # if isinstance(body, dict):
-        #     body = json.dumps(body)
+        if isinstance(body, dict):
+            body = json.dumps(body)
+        response = None
         try:
             print(f"Making API call: {method} {url} with body: {body}")
             if body == "{}":
                 body = None
-            response = requests.request(method, url, headers=headers, data=body)
+
+            proxy_id = self.config.get("proxy_id", None)
+            request_kwargs = {
+                'method': method,
+                'url': url,
+                'headers': sanitized_headers,
+                'data': body
+            }
+            try:
+                if proxy_id:
+                    from application.agents.tools.proxy_handler import apply_proxy_to_request
+                    response = apply_proxy_to_request(
+                        requests.request,
+                        proxy_id=proxy_id,
+                        **request_kwargs
+                    )
+                else:
+                    response = requests.request(**request_kwargs)
+            except ImportError:
+                response = requests.request(**request_kwargs)
             response.raise_for_status()
             content_type = response.headers.get(
                 "Content-Type", "application/json"

application/agents/tools/brave.py

@@ -25,35 +25,27 @@ class BraveSearchTool(Tool):
         else:
             raise ValueError(f"Unknown action: {action_name}")
 
-    def _web_search(
-        self,
-        query,
-        country="ALL",
-        search_lang="en",
-        count=10,
-        offset=0,
-        safesearch="off",
-        freshness=None,
-        result_filter=None,
-        extra_snippets=False,
-        summary=False,
-    ):
+    def _web_search(self, query, country="ALL", search_lang="en", count=10, 
+                   offset=0, safesearch="off", freshness=None, 
+                   result_filter=None, extra_snippets=False, summary=False):
         """
         Performs a web search using the Brave Search API.
         """
         print(f"Performing Brave web search for: {query}")
-
+        
         url = f"{self.base_url}/web/search"
-
+        
+        # Build query parameters
         params = {
             "q": query,
             "country": country,
             "search_lang": search_lang,
             "count": min(count, 20),
             "offset": min(offset, 9),
-            "safesearch": safesearch,
+            "safesearch": safesearch
         }
-
+        
+        # Add optional parameters only if they have values
         if freshness:
             params["freshness"] = freshness
         if result_filter:
@@ -62,69 +54,68 @@ class BraveSearchTool(Tool):
             params["extra_snippets"] = 1
         if summary:
             params["summary"] = 1
+        
+        # Set up headers
         headers = {
             "Accept": "application/json",
             "Accept-Encoding": "gzip",
-            "X-Subscription-Token": self.token,
+            "X-Subscription-Token": self.token
         }
-
+        
+        # Make the request
         response = requests.get(url, params=params, headers=headers)
-
+        
         if response.status_code == 200:
             return {
                 "status_code": response.status_code,
                 "results": response.json(),
-                "message": "Search completed successfully.",
+                "message": "Search completed successfully."
             }
         else:
             return {
                 "status_code": response.status_code,
-                "message": f"Search failed with status code: {response.status_code}.",
+                "message": f"Search failed with status code: {response.status_code}."
             }
-
-    def _image_search(
-        self,
-        query,
-        country="ALL",
-        search_lang="en",
-        count=5,
-        safesearch="off",
-        spellcheck=False,
-    ):
+    
+    def _image_search(self, query, country="ALL", search_lang="en", count=5, 
+                     safesearch="off", spellcheck=False):
         """
         Performs an image search using the Brave Search API.
         """
         print(f"Performing Brave image search for: {query}")
-
+        
         url = f"{self.base_url}/images/search"
-
+        
+        # Build query parameters
         params = {
             "q": query,
             "country": country,
             "search_lang": search_lang,
             "count": min(count, 100),  # API max is 100
             "safesearch": safesearch,
-            "spellcheck": 1 if spellcheck else 0,
+            "spellcheck": 1 if spellcheck else 0
         }
-
+        
+        # Set up headers
         headers = {
             "Accept": "application/json",
             "Accept-Encoding": "gzip",
-            "X-Subscription-Token": self.token,
+            "X-Subscription-Token": self.token
         }
-
+        
+        # Make the request
         response = requests.get(url, params=params, headers=headers)
-
+        
         if response.status_code == 200:
             return {
                 "status_code": response.status_code,
                 "results": response.json(),
-                "message": "Image search completed successfully.",
+                "message": "Image search completed successfully."
             }
         else:
             return {
                 "status_code": response.status_code,
-                "message": f"Image search failed with status code: {response.status_code}.",
+                "message": f"Image search failed with status code: {response.status_code}."
             }
 
     def get_actions_metadata(self):
@@ -139,14 +130,42 @@ class BraveSearchTool(Tool):
                             "type": "string",
                             "description": "The search query (max 400 characters, 50 words)",
                         },
+                        # "country": {
+                        #     "type": "string",
+                        #     "description": "The 2-character country code (default: US)",
+                        # },
                         "search_lang": {
                             "type": "string",
                             "description": "The search language preference (default: en)",
                         },
+                        # "count": {
+                        #     "type": "integer",
+                        #     "description": "Number of results to return (max 20, default: 10)",
+                        # },
+                        # "offset": {
+                        #     "type": "integer",
+                        #     "description": "Pagination offset (max 9, default: 0)",
+                        # },
+                        # "safesearch": {
+                        #     "type": "string",
+                        #     "description": "Filter level for adult content (off, moderate, strict)",
+                        # },
                         "freshness": {
                             "type": "string",
                             "description": "Time filter for results (pd: last 24h, pw: last week, pm: last month, py: last year)",
                         },
+                        # "result_filter": {
+                        #     "type": "string",
+                        #     "description": "Comma-delimited list of result types to include",
+                        # },
+                        # "extra_snippets": {
+                        #     "type": "boolean",
+                        #     "description": "Get additional excerpts from result pages",
+                        # },
+                        # "summary": {
+                        #     "type": "boolean",
+                        #     "description": "Enable summary generation in search results",
+                        # }
                     },
                     "required": ["query"],
                     "additionalProperties": False,
@@ -162,21 +181,37 @@ class BraveSearchTool(Tool):
                             "type": "string",
                             "description": "The search query (max 400 characters, 50 words)",
                         },
+                        # "country": {
+                        #     "type": "string",
+                        #     "description": "The 2-character country code (default: US)",
+                        # },
+                        # "search_lang": {
+                        #     "type": "string",
+                        #     "description": "The search language preference (default: en)",
+                        # },
                         "count": {
                             "type": "integer",
                             "description": "Number of results to return (max 100, default: 5)",
                         },
+                        # "safesearch": {
+                        #     "type": "string",
+                        #     "description": "Filter level for adult content (off, strict). Default: strict",
+                        # },
+                        # "spellcheck": {
+                        #     "type": "boolean",
+                        #     "description": "Whether to spellcheck provided query (default: true)",
+                        # }
                     },
                     "required": ["query"],
                     "additionalProperties": False,
                 },
-            },
+            }
         ]
 
     def get_config_requirements(self):
         return {
             "token": {
-                "type": "string",
-                "description": "Brave Search API key for authentication",
+                "type": "string", 
+                "description": "Brave Search API key for authentication"
             },
-        }
+        }

application/agents/tools/duckduckgo.py

@@ -1,114 +0,0 @@
-from application.agents.tools.base import Tool
-from duckduckgo_search import DDGS
-
-
-class DuckDuckGoSearchTool(Tool):
-    """
-    DuckDuckGo Search
-    A tool for performing web and image searches using DuckDuckGo.
-    """
-
-    def __init__(self, config):
-        self.config = config
-
-    def execute_action(self, action_name, **kwargs):
-        actions = {
-            "ddg_web_search": self._web_search,
-            "ddg_image_search": self._image_search,
-        }
-
-        if action_name in actions:
-            return actions[action_name](**kwargs)
-        else:
-            raise ValueError(f"Unknown action: {action_name}")
-
-    def _web_search(
-        self,
-        query,
-        max_results=5,
-    ):
-        print(f"Performing DuckDuckGo web search for: {query}")
-
-        try:
-            results = DDGS().text(
-                query,
-                max_results=max_results,
-            )
-
-            return {
-                "status_code": 200,
-                "results": results,
-                "message": "Web search completed successfully.",
-            }
-        except Exception as e:
-            return {
-                "status_code": 500,
-                "message": f"Web search failed: {str(e)}",
-            }
-
-    def _image_search(
-        self,
-        query,
-        max_results=5,
-    ):
-        print(f"Performing DuckDuckGo image search for: {query}")
-
-        try:
-            results = DDGS().images(
-                keywords=query,
-                max_results=max_results,
-            )
-
-            return {
-                "status_code": 200,
-                "results": results,
-                "message": "Image search completed successfully.",
-            }
-        except Exception as e:
-            return {
-                "status_code": 500,
-                "message": f"Image search failed: {str(e)}",
-            }
-
-    def get_actions_metadata(self):
-        return [
-            {
-                "name": "ddg_web_search",
-                "description": "Perform a web search using DuckDuckGo.",
-                "parameters": {
-                    "type": "object",
-                    "properties": {
-                        "query": {
-                            "type": "string",
-                            "description": "Search query",
-                        },
-                        "max_results": {
-                            "type": "integer",
-                            "description": "Number of results to return (default: 5)",
-                        },
-                    },
-                    "required": ["query"],
-                },
-            },
-            {
-                "name": "ddg_image_search",
-                "description": "Perform an image search using DuckDuckGo.",
-                "parameters": {
-                    "type": "object",
-                    "properties": {
-                        "query": {
-                            "type": "string",
-                            "description": "Search query",
-                        },
-                        "max_results": {
-                            "type": "integer",
-                            "description": "Number of results to return (default: 5, max: 50)",
-                        },
-                    },
-                    "required": ["query"],
-                },
-            },
-        ]
-
-    def get_config_requirements(self):
-        return {}

application/agents/tools/proxy_handler.py

@@ -0,0 +1,63 @@
+import logging
+import requests
+from typing import Dict, Optional
+from bson.objectid import ObjectId
+
+from application.core.mongo_db import MongoDB
+
+logger = logging.getLogger(__name__)
+
+# Get MongoDB connection
+mongo = MongoDB.get_client()
+db = mongo["docsgpt"]
+proxies_collection = db["proxies"]
+
+def get_proxy_config(proxy_id: str) -> Optional[Dict[str, str]]:
+    """
+    Retrieve proxy configuration from the database.
+    
+    Args:
+        proxy_id: The ID of the proxy configuration
+        
+    Returns:
+        A dictionary with proxy configuration or None if not found
+    """
+    if not proxy_id or proxy_id == "none":
+        return None
+        
+    try:
+        if ObjectId.is_valid(proxy_id):
+            proxy_config = proxies_collection.find_one({"_id": ObjectId(proxy_id)})
+            if proxy_config and "connection" in proxy_config:
+                connection_str = proxy_config["connection"].strip()
+                if connection_str:
+                    # Format proxy for requests library
+                    return {
+                        "http": connection_str,
+                        "https": connection_str
+                    }
+        return None
+    except Exception as e:
+        logger.error(f"Error retrieving proxy configuration: {e}")
+        return None
+
+def apply_proxy_to_request(request_func, proxy_id=None, **kwargs):
+    """
+    Apply proxy configuration to a requests function if available.
+    This is a minimal wrapper that doesn't change the function signature.
+    
+    Args:
+        request_func: The requests function to call (e.g., requests.get, requests.post)
+        proxy_id: Optional proxy ID to use
+        **kwargs: Arguments to pass to the request function
+    
+    Returns:
+        The response from the request
+    """
+    if proxy_id:
+        proxy_config = get_proxy_config(proxy_id)
+        if proxy_config:
+            kwargs['proxies'] = proxy_config
+            logger.info(f"Using proxy for request")
+    
+    return request_func(**kwargs)

application/agents/tools/read_webpage.py

@@ -1,83 +0,0 @@
-import requests
-from markdownify import markdownify
-from application.agents.tools.base import Tool
-from urllib.parse import urlparse
-
-class ReadWebpageTool(Tool):
-    """
-    Read Webpage (browser)
-    A tool to fetch the HTML content of a URL and convert it to Markdown.
-    """
-
-    def __init__(self, config=None):
-        """
-        Initializes the tool.
-        :param config: Optional configuration dictionary. Not used by this tool.
-        """
-        self.config = config
-
-    def execute_action(self, action_name: str, **kwargs) -> str:
-        """
-        Executes the specified action. For this tool, the only action is 'read_webpage'.
-
-        :param action_name: The name of the action to execute. Should be 'read_webpage'.
-        :param kwargs: Keyword arguments, must include 'url'.
-        :return: The Markdown content of the webpage or an error message.
-        """
-        if action_name != "read_webpage":
-            return f"Error: Unknown action '{action_name}'. This tool only supports 'read_webpage'."
-
-        url = kwargs.get("url")
-        if not url:
-            return "Error: URL parameter is missing."
-
-        # Ensure the URL has a scheme (if not, default to http)
-        parsed_url = urlparse(url)
-        if not parsed_url.scheme:
-            url = "http://" + url
-        
-        try:
-            response = requests.get(url, timeout=10, headers={'User-Agent': 'DocsGPT-Agent/1.0'})
-            response.raise_for_status()  # Raise an exception for HTTP errors (4xx or 5xx)
-            
-            html_content = response.text
-            #soup = BeautifulSoup(html_content, 'html.parser')
-            
-            
-            markdown_content = markdownify(html_content, heading_style="ATX", newline_style="BACKSLASH")
-            
-            return markdown_content
-
-        except requests.exceptions.RequestException as e:
-            return f"Error fetching URL {url}: {e}"
-        except Exception as e:
-            return f"Error processing URL {url}: {e}"
-
-    def get_actions_metadata(self):
-        """
-        Returns metadata for the actions supported by this tool.
-        """
-        return [
-            {
-                "name": "read_webpage",
-                "description": "Fetches the HTML content of a given URL and returns it as clean Markdown text. Input must be a valid URL.",
-                "parameters": {
-                    "type": "object",
-                    "properties": {
-                        "url": {
-                            "type": "string",
-                            "description": "The fully qualified URL of the webpage to read (e.g., 'https://www.example.com').",
-                        }
-                    },
-                    "required": ["url"],
-                    "additionalProperties": False,
-                },
-            }
-        ]
-
-    def get_config_requirements(self):
-        """
-        Returns a dictionary describing the configuration requirements for the tool.
-        This tool does not require any specific configuration.
-        """
-        return {}

application/agents/tools/tool_action_parser.py

@@ -17,21 +17,26 @@ class ToolActionParser:
         return parser(call)
 
     def _parse_openai_llm(self, call):
-        try:
-            call_args = json.loads(call.arguments)
-            tool_id = call.name.split("_")[-1]
-            action_name = call.name.rsplit("_", 1)[0]
-        except (AttributeError, TypeError) as e:
-            logger.error(f"Error parsing OpenAI LLM call: {e}")
-            return None, None, None
+        if isinstance(call, dict):
+            try:
+                call_args = json.loads(call["function"]["arguments"])
+                tool_id = call["function"]["name"].split("_")[-1]
+                action_name = call["function"]["name"].rsplit("_", 1)[0]
+            except (KeyError, TypeError) as e:
+                logger.error(f"Error parsing OpenAI LLM call: {e}")
+                return None, None, None
+        else:
+            try:
+                call_args = json.loads(call.function.arguments)
+                tool_id = call.function.name.split("_")[-1]
+                action_name = call.function.name.rsplit("_", 1)[0]
+            except (AttributeError, TypeError) as e:
+                logger.error(f"Error parsing OpenAI LLM call: {e}")
+                return None, None, None
         return tool_id, action_name, call_args
 
     def _parse_google_llm(self, call):
-        try:
-            call_args = call.arguments
-            tool_id = call.name.split("_")[-1]
-            action_name = call.name.rsplit("_", 1)[0]
-        except (AttributeError, TypeError) as e:
-            logger.error(f"Error parsing Google LLM call: {e}")
-            return None, None, None
+        call_args = call.args
+        tool_id = call.name.split("_")[-1]
+        action_name = call.name.rsplit("_", 1)[0]
         return tool_id, action_name, call_args

application/api/answer/routes.py

@@ -23,13 +23,12 @@ from application.utils import check_required_fields, limit_chat_history
 logger = logging.getLogger(__name__)
 
 mongo = MongoDB.get_client()
-db = mongo[settings.MONGO_DB_NAME]
+db = mongo["docsgpt"]
 conversations_collection = db["conversations"]
 sources_collection = db["sources"]
 prompts_collection = db["prompts"]
-agents_collection = db["agents"]
+api_key_collection = db["api_keys"]
 user_logs_collection = db["user_logs"]
-attachments_collection = db["attachments"]
 
 answer = Blueprint("answer", __name__)
 answer_ns = Namespace("answer", description="Answer related operations", path="/")
@@ -37,17 +36,17 @@ api.add_namespace(answer_ns)
 
 gpt_model = ""
 # to have some kind of default behaviour
-if settings.LLM_PROVIDER == "openai":
+if settings.LLM_NAME == "openai":
     gpt_model = "gpt-4o-mini"
-elif settings.LLM_PROVIDER == "anthropic":
+elif settings.LLM_NAME == "anthropic":
     gpt_model = "claude-2"
-elif settings.LLM_PROVIDER == "groq":
+elif settings.LLM_NAME == "groq":
     gpt_model = "llama3-8b-8192"
-elif settings.LLM_PROVIDER == "novita":
+elif settings.LLM_NAME == "novita":
     gpt_model = "deepseek/deepseek-r1"
 
-if settings.LLM_NAME:  # in case there is particular model name configured
-    gpt_model = settings.LLM_NAME
+if settings.MODEL_NAME:  # in case there is particular model name configured
+    gpt_model = settings.MODEL_NAME
 
 # load the prompts
 current_dir = os.path.dirname(
@@ -86,51 +85,19 @@ def run_async_chain(chain, question, chat_history):
     return result
 
 
-def get_agent_key(agent_id, user_id):
-    if not agent_id:
-        return None, False, None
-
-    try:
-        agent = agents_collection.find_one({"_id": ObjectId(agent_id)})
-        if agent is None:
-            raise Exception("Agent not found", 404)
-
-        is_owner = agent.get("user") == user_id
-
-        if is_owner:
-            agents_collection.update_one(
-                {"_id": ObjectId(agent_id)},
-                {"$set": {"lastUsedAt": datetime.datetime.now(datetime.timezone.utc)}},
-            )
-            return str(agent["key"]), False, None
-
-        is_shared_with_user = agent.get(
-            "shared_publicly", False
-        ) or user_id in agent.get("shared_with", [])
-
-        if is_shared_with_user:
-            return str(agent["key"]), True, agent.get("shared_token")
-
-        raise Exception("Unauthorized access to the agent", 403)
-
-    except Exception as e:
-        logger.error(f"Error in get_agent_key: {str(e)}", exc_info=True)
-        raise
-
-
 def get_data_from_api_key(api_key):
-    data = agents_collection.find_one({"key": api_key})
-    if not data:
-        raise Exception("Invalid API Key, please generate a new key", 401)
+    data = api_key_collection.find_one({"key": api_key})
+    # # Raise custom exception if the API key is not found
+    if data is None:
+        raise Exception("Invalid API Key, please generate new key", 401)
 
-    source = data.get("source")
-    if isinstance(source, DBRef):
-        source_doc = db.dereference(source)
+    if "source" in data and isinstance(data["source"], DBRef):
+        source_doc = db.dereference(data["source"])
         data["source"] = str(source_doc["_id"])
-        data["retriever"] = source_doc.get("retriever", data.get("retriever"))
+        if "retriever" in source_doc:
+            data["retriever"] = source_doc["retriever"]
     else:
         data["source"] = {}
-
     return data
 
 
@@ -154,17 +121,12 @@ def save_conversation(
     conversation_id,
     question,
     response,
-    thought,
     source_log_docs,
     tool_calls,
     llm,
     decoded_token,
     index=None,
     api_key=None,
-    agent_id=None,
-    is_shared_usage=False,
-    shared_token=None,
-    attachment_ids=None,
 ):
     current_time = datetime.datetime.now(datetime.timezone.utc)
     if conversation_id is not None and index is not None:
@@ -174,11 +136,9 @@ def save_conversation(
                 "$set": {
                     f"queries.{index}.prompt": question,
                     f"queries.{index}.response": response,
-                    f"queries.{index}.thought": thought,
                     f"queries.{index}.sources": source_log_docs,
                     f"queries.{index}.tool_calls": tool_calls,
                     f"queries.{index}.timestamp": current_time,
-                    f"queries.{index}.attachments": attachment_ids,
                 }
             },
         )
@@ -195,11 +155,9 @@ def save_conversation(
                     "queries": {
                         "prompt": question,
                         "response": response,
-                        "thought": thought,
                         "sources": source_log_docs,
                         "tool_calls": tool_calls,
                         "timestamp": current_time,
-                        "attachments": attachment_ids,
                     }
                 }
             },
@@ -213,13 +171,13 @@ def save_conversation(
                 "role": "assistant",
                 "content": "Summarise following conversation in no more than 3 "
                 "words, respond ONLY with the summary, use the same "
-                "language as the user query",
+                "language as the system",
             },
             {
                 "role": "user",
                 "content": "Summarise following conversation in no more than 3 words, "
                 "respond ONLY with the summary, use the same language as the "
-                "user query \n\nUser: " + question + "\n\n" + "AI: " + response,
+                "system \n\nUser: " + question + "\n\n" + "AI: " + response,
             },
         ]
 
@@ -232,21 +190,14 @@ def save_conversation(
                 {
                     "prompt": question,
                     "response": response,
-                    "thought": thought,
                     "sources": source_log_docs,
                     "tool_calls": tool_calls,
                     "timestamp": current_time,
-                    "attachments": attachment_ids,
                 }
             ],
         }
         if api_key:
-            if agent_id:
-                conversation_data["agent_id"] = agent_id
-                if is_shared_usage:
-                    conversation_data["is_shared_usage"] = is_shared_usage
-                    conversation_data["shared_token"] = shared_token
-            api_key_doc = agents_collection.find_one({"key": api_key})
+            api_key_doc = api_key_collection.find_one({"key": api_key})
             if api_key_doc:
                 conversation_data["api_key"] = api_key_doc["key"]
         conversation_id = conversations_collection.insert_one(
@@ -277,13 +228,11 @@ def complete_stream(
     isNoneDoc=False,
     index=None,
     should_save_conversation=True,
-    attachment_ids=None,
-    agent_id=None,
-    is_shared_usage=False,
-    shared_token=None,
 ):
     try:
-        response_full, thought, source_log_docs, tool_calls = "", "", [], []
+        response_full = ""
+        source_log_docs = []
+        tool_calls = []
 
         answer = agent.gen(query=question, retriever=retriever)
 
@@ -307,12 +256,7 @@ def complete_stream(
                     yield f"data: {data}\n\n"
             elif "tool_calls" in line:
                 tool_calls = line["tool_calls"]
-            elif "thought" in line:
-                thought += line["thought"]
-                data = json.dumps({"type": "thought", "thought": line["thought"]})
-                yield f"data: {data}\n\n"
-            elif "type" in line:
-                data = json.dumps(line)
+                data = json.dumps({"type": "tool_calls", "tool_calls": tool_calls})
                 yield f"data: {data}\n\n"
 
         if isNoneDoc:
@@ -320,7 +264,7 @@ def complete_stream(
                 doc["source"] = "None"
 
         llm = LLMCreator.create_llm(
-            settings.LLM_PROVIDER,
+            settings.LLM_NAME,
             api_key=settings.API_KEY,
             user_api_key=user_api_key,
             decoded_token=decoded_token,
@@ -331,17 +275,12 @@ def complete_stream(
                 conversation_id,
                 question,
                 response_full,
-                thought,
                 source_log_docs,
                 tool_calls,
                 llm,
                 decoded_token,
                 index,
                 api_key=user_api_key,
-                attachment_ids=attachment_ids,
-                agent_id=agent_id,
-                is_shared_usage=is_shared_usage,
-                shared_token=shared_token,
             )
         else:
             conversation_id = None
@@ -361,14 +300,14 @@ def complete_stream(
                 "response": response_full,
                 "sources": source_log_docs,
                 "retriever_params": retriever_params,
-                "attachments": attachment_ids,
                 "timestamp": datetime.datetime.now(datetime.timezone.utc),
             }
         )
         data = json.dumps({"type": "end"})
         yield f"data: {data}\n\n"
     except Exception as e:
-        logger.error(f"Error in stream: {str(e)}", exc_info=True)
+        logger.error(f"Error in stream: {str(e)}")
+        logger.error(traceback.format_exc())
         data = json.dumps(
             {
                 "type": "error",
@@ -396,6 +335,9 @@ class Stream(Resource):
             "prompt_id": fields.String(
                 required=False, default="default", description="Prompt ID"
             ),
+            "proxy_id": fields.String(
+                required=False, description="Proxy ID to use for API calls"
+            ),
             "chunks": fields.Integer(
                 required=False, default=2, description="Number of chunks"
             ),
@@ -409,15 +351,10 @@ class Stream(Resource):
                 required=False, description="Flag indicating if no document is used"
             ),
             "index": fields.Integer(
-                required=False, description="Index of the query to update"
+                required=False, description="The position where query is to be updated"
             ),
             "save_conversation": fields.Boolean(
-                required=False,
-                default=True,
-                description="Whether to save the conversation",
-            ),
-            "attachments": fields.List(
-                fields.String, required=False, description="List of attachment IDs"
+                required=False, default=True, description="Flag to save conversation"
             ),
         },
     )
@@ -438,40 +375,26 @@ class Stream(Resource):
         try:
             question = data["question"]
             history = limit_chat_history(
-                json.loads(data.get("history", "[]")), gpt_model=gpt_model
+                json.loads(data.get("history", [])), gpt_model=gpt_model
             )
             conversation_id = data.get("conversation_id")
             prompt_id = data.get("prompt_id", "default")
-            attachment_ids = data.get("attachments", [])
+            proxy_id = data.get("proxy_id", None)
 
             index = data.get("index", None)
             chunks = int(data.get("chunks", 2))
             token_limit = data.get("token_limit", settings.DEFAULT_MAX_HISTORY)
             retriever_name = data.get("retriever", "classic")
-            agent_id = data.get("agent_id", None)
-            agent_type = settings.AGENT_NAME
-            decoded_token = getattr(request, "decoded_token", None)
-            user_sub = decoded_token.get("sub") if decoded_token else None
-            agent_key, is_shared_usage, shared_token = get_agent_key(agent_id, user_sub)
-
-            if agent_key:
-                data.update({"api_key": agent_key})
-            else:
-                agent_id = None
 
             if "api_key" in data:
                 data_key = get_data_from_api_key(data["api_key"])
                 chunks = int(data_key.get("chunks", 2))
                 prompt_id = data_key.get("prompt_id", "default")
+                proxy_id = data_key.get("proxy_id", None)
                 source = {"active_docs": data_key.get("source")}
                 retriever_name = data_key.get("retriever", retriever_name)
                 user_api_key = data["api_key"]
-                agent_type = data_key.get("agent_type", agent_type)
-                if is_shared_usage:
-                    decoded_token = request.decoded_token
-                else:
-                    decoded_token = {"sub": data_key.get("user")}
-                    is_shared_usage = False
+                decoded_token = {"sub": data_key.get("user")}
 
             elif "active_docs" in data:
                 source = {"active_docs": data["active_docs"]}
@@ -487,12 +410,8 @@ class Stream(Resource):
             if not decoded_token:
                 return make_response({"error": "Unauthorized"}, 401)
 
-            attachments = get_attachments_content(
-                attachment_ids, decoded_token.get("sub")
-            )
-
             logger.info(
-                f"/stream - request_data: {data}, source: {source}, attachments: {len(attachments)}",
+                f"/stream - request_data: {data}, source: {source}",
                 extra={"data": json.dumps({"request_data": data, "source": source})},
             )
 
@@ -501,16 +420,16 @@ class Stream(Resource):
                 chunks = 0
 
             agent = AgentCreator.create_agent(
-                agent_type,
+                settings.AGENT_NAME,
                 endpoint="stream",
-                llm_name=settings.LLM_PROVIDER,
+                llm_name=settings.LLM_NAME,
                 gpt_model=gpt_model,
                 api_key=settings.API_KEY,
                 user_api_key=user_api_key,
                 prompt=prompt,
+                proxy_id=proxy_id,
                 chat_history=history,
                 decoded_token=decoded_token,
-                attachments=attachments,
             )
 
             retriever = RetrieverCreator.create_retriever(
@@ -536,10 +455,6 @@ class Stream(Resource):
                     isNoneDoc=data.get("isNoneDoc"),
                     index=index,
                     should_save_conversation=save_conv,
-                    attachment_ids=attachment_ids,
-                    agent_id=agent_id,
-                    is_shared_usage=is_shared_usage,
-                    shared_token=shared_token,
                 ),
                 mimetype="text/event-stream",
             )
@@ -587,6 +502,9 @@ class Answer(Resource):
             "prompt_id": fields.String(
                 required=False, default="default", description="Prompt ID"
             ),
+            "proxy_id": fields.String(
+                required=False, description="Proxy ID to use for API calls"
+            ),
             "chunks": fields.Integer(
                 required=False, default=2, description="Number of chunks"
             ),
@@ -614,23 +532,23 @@ class Answer(Resource):
         try:
             question = data["question"]
             history = limit_chat_history(
-                json.loads(data.get("history", "[]")), gpt_model=gpt_model
+                json.loads(data.get("history", [])), gpt_model=gpt_model
             )
             conversation_id = data.get("conversation_id")
             prompt_id = data.get("prompt_id", "default")
+            proxy_id = data.get("proxy_id", None)
             chunks = int(data.get("chunks", 2))
             token_limit = data.get("token_limit", settings.DEFAULT_MAX_HISTORY)
             retriever_name = data.get("retriever", "classic")
-            agent_type = settings.AGENT_NAME
 
             if "api_key" in data:
                 data_key = get_data_from_api_key(data["api_key"])
                 chunks = int(data_key.get("chunks", 2))
                 prompt_id = data_key.get("prompt_id", "default")
+                proxy_id = data_key.get("proxy_id", None)
                 source = {"active_docs": data_key.get("source")}
                 retriever_name = data_key.get("retriever", retriever_name)
                 user_api_key = data["api_key"]
-                agent_type = data_key.get("agent_type", agent_type)
                 decoded_token = {"sub": data_key.get("user")}
 
             elif "active_docs" in data:
@@ -655,13 +573,14 @@ class Answer(Resource):
             )
 
             agent = AgentCreator.create_agent(
-                agent_type,
+                settings.AGENT_NAME,
                 endpoint="api/answer",
-                llm_name=settings.LLM_PROVIDER,
+                llm_name=settings.LLM_NAME,
                 gpt_model=gpt_model,
                 api_key=settings.API_KEY,
                 user_api_key=user_api_key,
                 prompt=prompt,
+                proxy_id=proxy_id,
                 chat_history=history,
                 decoded_token=decoded_token,
             )
@@ -682,7 +601,6 @@ class Answer(Resource):
             source_log_docs = []
             tool_calls = []
             stream_ended = False
-            thought = ""
 
             for line in complete_stream(
                 question=question,
@@ -705,8 +623,6 @@ class Answer(Resource):
                         source_log_docs = event["source"]
                     elif event["type"] == "tool_calls":
                         tool_calls = event["tool_calls"]
-                    elif event["type"] == "thought":
-                        thought = event["thought"]
                     elif event["type"] == "error":
                         logger.error(f"Error from stream: {event['error']}")
                         return bad_request(500, event["error"])
@@ -726,7 +642,7 @@ class Answer(Resource):
                     doc["source"] = "None"
 
             llm = LLMCreator.create_llm(
-                settings.LLM_PROVIDER,
+                settings.LLM_NAME,
                 api_key=settings.API_KEY,
                 user_api_key=user_api_key,
                 decoded_token=decoded_token,
@@ -738,7 +654,6 @@ class Answer(Resource):
                     conversation_id,
                     question,
                     response_full,
-                    thought,
                     source_log_docs,
                     tool_calls,
                     llm,
@@ -881,34 +796,3 @@ class Search(Resource):
             return bad_request(500, str(e))
 
         return make_response(docs, 200)
-
-
-def get_attachments_content(attachment_ids, user):
-    """
-    Retrieve content from attachment documents based on their IDs.
-
-    Args:
-        attachment_ids (list): List of attachment document IDs
-        user (str): User identifier to verify ownership
-
-    Returns:
-        list: List of dictionaries containing attachment content and metadata
-    """
-    if not attachment_ids:
-        return []
-
-    attachments = []
-    for attachment_id in attachment_ids:
-        try:
-            attachment_doc = attachments_collection.find_one(
-                {"_id": ObjectId(attachment_id), "user": user}
-            )
-
-            if attachment_doc:
-                attachments.append(attachment_doc)
-        except Exception as e:
-            logger.error(
-                f"Error retrieving attachment {attachment_id}: {e}", exc_info=True
-            )
-
-    return attachments

application/api/internal/routes.py

@@ -3,15 +3,12 @@ import datetime
 from flask import Blueprint, request, send_from_directory
 from werkzeug.utils import secure_filename
 from bson.objectid import ObjectId
-import logging
+
 from application.core.mongo_db import MongoDB
 from application.core.settings import settings
-from application.storage.storage_creator import StorageCreator
 
-
-logger = logging.getLogger(__name__)
 mongo = MongoDB.get_client()
-db = mongo[settings.MONGO_DB_NAME]
+db = mongo["docsgpt"]
 conversations_collection = db["conversations"]
 sources_collection = db["sources"]
 
@@ -37,39 +34,37 @@ def upload_index_files():
     """Upload two files(index.faiss, index.pkl) to the user's folder."""
     if "user" not in request.form:
         return {"status": "no user"}
-    user = request.form["user"] 
+    user = secure_filename(request.form["user"])
     if "name" not in request.form:
         return {"status": "no name"}
-    job_name = request.form["name"]
-    tokens = request.form["tokens"]
-    retriever = request.form["retriever"]
-    id = request.form["id"]
-    type = request.form["type"]
+    job_name = secure_filename(request.form["name"])
+    tokens = secure_filename(request.form["tokens"])
+    retriever = secure_filename(request.form["retriever"])
+    id = secure_filename(request.form["id"])
+    type = secure_filename(request.form["type"])
     remote_data = request.form["remote_data"] if "remote_data" in request.form else None
-    sync_frequency = request.form["sync_frequency"] if "sync_frequency" in request.form else None
-    
-    original_file_path = request.form.get("original_file_path")
+    sync_frequency = secure_filename(request.form["sync_frequency"]) if "sync_frequency" in request.form else None
 
-    storage = StorageCreator.get_storage()
-    index_base_path = f"indexes/{id}"
-    
+    save_dir = os.path.join(current_dir, "indexes", str(id))
     if settings.VECTOR_STORE == "faiss":
         if "file_faiss" not in request.files:
-            logger.error("No file_faiss part")
+            print("No file part")
             return {"status": "no file"}
         file_faiss = request.files["file_faiss"]
         if file_faiss.filename == "":
             return {"status": "no file name"}
         if "file_pkl" not in request.files:
-            logger.error("No file_pkl part")
+            print("No file part")
             return {"status": "no file"}
         file_pkl = request.files["file_pkl"]
         if file_pkl.filename == "":
             return {"status": "no file name"}
-        
-        # Save index files to storage
-        storage.save_file(file_faiss, f"{index_base_path}/index.faiss")
-        storage.save_file(file_pkl, f"{index_base_path}/index.pkl")
+        # saves index files
+
+        if not os.path.exists(save_dir):
+            os.makedirs(save_dir)
+        file_faiss.save(os.path.join(save_dir, "index.faiss"))
+        file_pkl.save(os.path.join(save_dir, "index.pkl"))
 
     existing_entry = sources_collection.find_one({"_id": ObjectId(id)})
     if existing_entry:
@@ -87,7 +82,6 @@ def upload_index_files():
                     "retriever": retriever,
                     "remote_data": remote_data,
                     "sync_frequency": sync_frequency,
-                    "file_path": original_file_path,
                 }
             },
         )
@@ -105,7 +99,6 @@ def upload_index_files():
                 "retriever": retriever,
                 "remote_data": remote_data,
                 "sync_frequency": sync_frequency,
-                "file_path": original_file_path,
             }
         )
     return {"status": "ok"}

application/api/user/tasks.py

@@ -1,18 +1,12 @@
 from datetime import timedelta
 
 from application.celery_init import celery
-from application.worker import (
-    agent_webhook_worker,
-    attachment_worker,
-    ingest_worker,
-    remote_worker,
-    sync_worker,
-)
+from application.worker import ingest_worker, remote_worker, sync_worker
 
 
 @celery.task(bind=True)
-def ingest(self, directory, formats, job_name, filename, user, dir_name, user_dir):
-    resp = ingest_worker(self, directory, formats, job_name, filename, user, dir_name, user_dir)
+def ingest(self, directory, formats, name_job, filename, user):
+    resp = ingest_worker(self, directory, formats, name_job, filename, user)
     return resp
 
 
@@ -28,18 +22,6 @@ def schedule_syncs(self, frequency):
     return resp
 
 
-@celery.task(bind=True)
-def store_attachment(self, file_info, user):
-    resp = attachment_worker(self, file_info, user)
-    return resp
-
-
-@celery.task(bind=True)
-def process_agent_webhook(self, agent_id, payload):
-    resp = agent_webhook_worker(self, agent_id, payload)
-    return resp
-
-
 @celery.on_after_configure.connect
 def setup_periodic_tasks(sender, **kwargs):
     sender.add_periodic_task(

application/cache.py

@@ -61,14 +61,14 @@ def gen_cache(func):
                 if cached_response:
                     return cached_response.decode("utf-8")
             except Exception as e:
-                logger.error(f"Error getting cached response: {e}", exc_info=True)
+                logger.error(f"Error getting cached response: {e}")
 
         result = func(self, model, messages, stream, tools, *args, **kwargs)
         if redis_client and isinstance(result, str):
             try:
                 redis_client.set(cache_key, result, ex=1800)
             except Exception as e:
-                logger.error(f"Error setting cache: {e}", exc_info=True)
+                logger.error(f"Error setting cache: {e}")
 
         return result
 
@@ -100,7 +100,7 @@ def stream_cache(func):
                         time.sleep(0.03)  # Simulate streaming delay
                     return
             except Exception as e:
-                logger.error(f"Error getting cached stream: {e}", exc_info=True)
+                logger.error(f"Error getting cached stream: {e}")
 
         stream_cache_data = []
         for chunk in func(self, model, messages, stream, tools, *args, **kwargs):
@@ -112,6 +112,6 @@ def stream_cache(func):
                 redis_client.set(cache_key, json.dumps(stream_cache_data), ex=1800)
                 logger.info(f"Stream cache saved for key: {cache_key}")
             except Exception as e:
-                logger.error(f"Error setting stream cache: {e}", exc_info=True)
+                logger.error(f"Error setting stream cache: {e}")
 
     return wrapper

application/core/settings.py

@@ -11,18 +11,17 @@ current_dir = os.path.dirname(
 
 class Settings(BaseSettings):
     AUTH_TYPE: Optional[str] = None
-    LLM_PROVIDER: str = "docsgpt"
-    LLM_NAME: Optional[str] = (
-        None  # if LLM_PROVIDER is openai, LLM_NAME can be gpt-4 or gpt-3.5-turbo
+    LLM_NAME: str = "docsgpt"
+    MODEL_NAME: Optional[str] = (
+        None  # if LLM_NAME is openai, MODEL_NAME can be gpt-4 or gpt-3.5-turbo
     )
     EMBEDDINGS_NAME: str = "huggingface_sentence-transformers/all-mpnet-base-v2"
     CELERY_BROKER_URL: str = "redis://localhost:6379/0"
     CELERY_RESULT_BACKEND: str = "redis://localhost:6379/1"
     MONGO_URI: str = "mongodb://localhost:27017/docsgpt"
-    MONGO_DB_NAME: str = "docsgpt"
-    LLM_PATH: str = os.path.join(current_dir, "models/docsgpt-7b-f16.gguf")
+    MODEL_PATH: str = os.path.join(current_dir, "models/docsgpt-7b-f16.gguf")
     DEFAULT_MAX_HISTORY: int = 150
-    LLM_TOKEN_LIMITS: dict = {
+    MODEL_TOKEN_LIMITS: dict = {
         "gpt-4o-mini": 128000,
         "gpt-3.5-turbo": 4096,
         "claude-2": 1e5,
@@ -33,11 +32,8 @@ class Settings(BaseSettings):
     VECTOR_STORE: str = (
         "faiss"  #  "faiss" or "elasticsearch" or "qdrant" or "milvus" or "lancedb"
     )
-    RETRIEVERS_ENABLED: list = ["classic_rag"]
+    RETRIEVERS_ENABLED: list = ["classic_rag", "duckduck_search"]  # also brave_search
     AGENT_NAME: str = "classic"
-    FALLBACK_LLM_PROVIDER: Optional[str] = None  # provider for fallback llm
-    FALLBACK_LLM_NAME: Optional[str] = None  # model name for fallback llm
-    FALLBACK_LLM_API_KEY: Optional[str] = None  # api key for fallback llm
 
     # LLM Cache
     CACHE_REDIS_URL: str = "redis://localhost:6379/2"
@@ -99,10 +95,9 @@ class Settings(BaseSettings):
     LANCEDB_TABLE_NAME: Optional[str] = (
         "docsgpts"  # Name of the table to use for storing vectors
     )
+    BRAVE_SEARCH_API_KEY: Optional[str] = None
 
     FLASK_DEBUG_MODE: bool = False
-    STORAGE_TYPE: str = "local"  # local or s3
-    URL_STRATEGY: str = "backend"  # backend or s3
 
     JWT_SECRET_KEY: str = ""
 

application/llm/base.py

@@ -1,117 +1,53 @@
-import logging
 from abc import ABC, abstractmethod
 
 from application.cache import gen_cache, stream_cache
-
-from application.core.settings import settings
 from application.usage import gen_token_usage, stream_token_usage
 
-logger = logging.getLogger(__name__)
-
 
 class BaseLLM(ABC):
-    def __init__(
-        self,
-        decoded_token=None,
-    ):
+    def __init__(self, decoded_token=None):
         self.decoded_token = decoded_token
         self.token_usage = {"prompt_tokens": 0, "generated_tokens": 0}
-        self.fallback_provider = settings.FALLBACK_LLM_PROVIDER
-        self.fallback_model_name = settings.FALLBACK_LLM_NAME
-        self.fallback_llm_api_key = settings.FALLBACK_LLM_API_KEY
-        self._fallback_llm = None
 
-    @property
-    def fallback_llm(self):
-        """Lazy-loaded fallback LLM instance."""
-        if (
-            self._fallback_llm is None
-            and self.fallback_provider
-            and self.fallback_model_name
-        ):
-            try:
-                from application.llm.llm_creator import LLMCreator
-
-                self._fallback_llm = LLMCreator.create_llm(
-                    self.fallback_provider,
-                    self.fallback_llm_api_key,
-                    None,
-                    self.decoded_token,
-                )
-            except Exception as e:
-                logger.error(
-                    f"Failed to initialize fallback LLM: {str(e)}", exc_info=True
-                )
-        return self._fallback_llm
-
-    def _execute_with_fallback(
-        self, method_name: str, decorators: list, *args, **kwargs
-    ):
-        """
-        Unified method execution with fallback support.
-
-        Args:
-            method_name: Name of the raw method ('_raw_gen' or '_raw_gen_stream')
-            decorators: List of decorators to apply
-            *args: Positional arguments
-            **kwargs: Keyword arguments
-        """
-
-        def decorated_method():
-            method = getattr(self, method_name)
-            for decorator in decorators:
-                method = decorator(method)
-            return method(self, *args, **kwargs)
-
-        try:
-            return decorated_method()
-        except Exception as e:
-            if not self.fallback_llm:
-                logger.error(f"Primary LLM failed and no fallback available: {str(e)}")
-                raise
-            logger.warning(
-                f"Falling back to {self.fallback_provider}/{self.fallback_model_name}. Error: {str(e)}"
-            )
-
-            fallback_method = getattr(
-                self.fallback_llm, method_name.replace("_raw_", "")
-            )
-            return fallback_method(*args, **kwargs)
-
-    def gen(self, model, messages, stream=False, tools=None, *args, **kwargs):
-        decorators = [gen_token_usage, gen_cache]
-        return self._execute_with_fallback(
-            "_raw_gen",
-            decorators,
-            model=model,
-            messages=messages,
-            stream=stream,
-            tools=tools,
-            *args,
-            **kwargs,
-        )
-
-    def gen_stream(self, model, messages, stream=True, tools=None, *args, **kwargs):
-        decorators = [stream_cache, stream_token_usage]
-        return self._execute_with_fallback(
-            "_raw_gen_stream",
-            decorators,
-            model=model,
-            messages=messages,
-            stream=stream,
-            tools=tools,
-            *args,
-            **kwargs,
-        )
+    def _apply_decorator(self, method, decorators, *args, **kwargs):
+        for decorator in decorators:
+            method = decorator(method)
+        return method(self, *args, **kwargs)
 
     @abstractmethod
     def _raw_gen(self, model, messages, stream, tools, *args, **kwargs):
         pass
 
+    def gen(self, model, messages, stream=False, tools=None, *args, **kwargs):
+        decorators = [gen_token_usage, gen_cache]
+        return self._apply_decorator(
+            self._raw_gen,
+            decorators=decorators,
+            model=model,
+            messages=messages,
+            stream=stream,
+            tools=tools,
+            *args,
+            **kwargs
+        )
+
     @abstractmethod
     def _raw_gen_stream(self, model, messages, stream, *args, **kwargs):
         pass
 
+    def gen_stream(self, model, messages, stream=True, tools=None, *args, **kwargs):
+        decorators = [stream_cache, stream_token_usage]
+        return self._apply_decorator(
+            self._raw_gen_stream,
+            decorators=decorators,
+            model=model,
+            messages=messages,
+            stream=stream,
+            tools=tools,
+            *args,
+            **kwargs
+        )
+
     def supports_tools(self):
         return hasattr(self, "_supports_tools") and callable(
             getattr(self, "_supports_tools")
@@ -119,12 +55,3 @@ class BaseLLM(ABC):
 
     def _supports_tools(self):
         raise NotImplementedError("Subclass must implement _supports_tools method")
-
-    def get_supported_attachment_types(self):
-        """
-        Return a list of MIME types supported by this LLM for file uploads.
-
-        Returns:
-            list: List of supported MIME types
-        """
-        return []  # Default: no attachments supported

application/llm/google_ai.py

@@ -1,11 +1,7 @@
 from google import genai
 from google.genai import types
-import logging
-import json
 
 from application.llm.base import BaseLLM
-from application.storage.storage_creator import StorageCreator
-from application.core.settings import settings
 
 
 class GoogleLLM(BaseLLM):
@@ -13,126 +9,6 @@ class GoogleLLM(BaseLLM):
         super().__init__(*args, **kwargs)
         self.api_key = api_key
         self.user_api_key = user_api_key
-        self.client = genai.Client(api_key=self.api_key)
-        self.storage = StorageCreator.get_storage()
-
-    def get_supported_attachment_types(self):
-        """
-        Return a list of MIME types supported by Google Gemini for file uploads.
-
-        Returns:
-            list: List of supported MIME types
-        """
-        return [
-            'application/pdf',
-            'image/png',
-            'image/jpeg',
-            'image/jpg',
-            'image/webp',
-            'image/gif'
-        ]
-
-    def prepare_messages_with_attachments(self, messages, attachments=None):
-        """
-        Process attachments using Google AI's file API for more efficient handling.
-
-        Args:
-            messages (list): List of message dictionaries.
-            attachments (list): List of attachment dictionaries with content and metadata.
-
-        Returns:
-            list: Messages formatted with file references for Google AI API.
-        """
-        if not attachments:
-            return messages
-
-        prepared_messages = messages.copy()
-
-        # Find the user message to attach files to the last one
-        user_message_index = None
-        for i in range(len(prepared_messages) - 1, -1, -1):
-            if prepared_messages[i].get("role") == "user":
-                user_message_index = i
-                break
-
-        if user_message_index is None:
-            user_message = {"role": "user", "content": []}
-            prepared_messages.append(user_message)
-            user_message_index = len(prepared_messages) - 1
-
-        if isinstance(prepared_messages[user_message_index].get("content"), str):
-            text_content = prepared_messages[user_message_index]["content"]
-            prepared_messages[user_message_index]["content"] = [
-                {"type": "text", "text": text_content}
-            ]
-        elif not isinstance(prepared_messages[user_message_index].get("content"), list):
-            prepared_messages[user_message_index]["content"] = []
-
-        files = []
-        for attachment in attachments:
-            mime_type = attachment.get('mime_type')
-
-            if mime_type in self.get_supported_attachment_types():
-                try:
-                    file_uri = self._upload_file_to_google(attachment)
-                    logging.info(f"GoogleLLM: Successfully uploaded file, got URI: {file_uri}")
-                    files.append({"file_uri": file_uri, "mime_type": mime_type})
-                except Exception as e:
-                    logging.error(f"GoogleLLM: Error uploading file: {e}", exc_info=True)
-                    if 'content' in attachment:
-                        prepared_messages[user_message_index]["content"].append({
-                            "type": "text",
-                            "text": f"[File could not be processed: {attachment.get('path', 'unknown')}]"
-                        })
-
-        if files:
-            logging.info(f"GoogleLLM: Adding {len(files)} files to message")
-            prepared_messages[user_message_index]["content"].append({
-                "files": files
-            })
-
-        return prepared_messages
-
-    def _upload_file_to_google(self, attachment):
-        """
-        Upload a file to Google AI and return the file URI.
-
-        Args:
-            attachment (dict): Attachment dictionary with path and metadata.
-
-        Returns:
-            str: Google AI file URI for the uploaded file.
-        """
-        if 'google_file_uri' in attachment:
-            return attachment['google_file_uri']
-
-        file_path = attachment.get('path')
-        if not file_path:
-            raise ValueError("No file path provided in attachment")
-
-        if not self.storage.file_exists(file_path):
-            raise FileNotFoundError(f"File not found: {file_path}")
-
-        try:
-            file_uri = self.storage.process_file(
-                file_path,
-                lambda local_path, **kwargs: self.client.files.upload(file=local_path).uri
-            )
-
-            from application.core.mongo_db import MongoDB
-            mongo = MongoDB.get_client()
-            db = mongo[settings.MONGO_DB_NAME]
-            attachments_collection = db["attachments"]
-            if '_id' in attachment:
-                attachments_collection.update_one(
-                    {"_id": attachment['_id']},
-                    {"$set": {"google_file_uri": file_uri}}
-                )
-
-            return file_uri
-        except Exception as e:
-            logging.error(f"Error uploading file to Google AI: {e}", exc_info=True)
-            raise
 
     def _clean_messages_google(self, messages):
         cleaned_messages = []
@@ -150,7 +26,7 @@ class GoogleLLM(BaseLLM):
                 elif isinstance(content, list):
                     for item in content:
                         if "text" in item:
-                            parts.append(types.Part.from_text(text=item["text"]))
+                            parts.append(types.Part.from_text(item["text"]))
                         elif "function_call" in item:
                             parts.append(
                                 types.Part.from_function_call(
@@ -165,14 +41,6 @@ class GoogleLLM(BaseLLM):
                                     response=item["function_response"]["response"],
                                 )
                             )
-                        elif "files" in item:
-                                for file_data in item["files"]:
-                                    parts.append(
-                                        types.Part.from_uri(
-                                            file_uri=file_data["file_uri"],
-                                            mime_type=file_data["mime_type"]
-                                        )
-                                    )
                         else:
                             raise ValueError(
                                 f"Unexpected content dictionary format:{item}"
@@ -278,25 +146,11 @@ class GoogleLLM(BaseLLM):
             cleaned_tools = self._clean_tools_format(tools)
             config.tools = cleaned_tools
 
-        # Check if we have both tools and file attachments
-        has_attachments = False
-        for message in messages:
-            for part in message.parts:
-                if hasattr(part, 'file_data') and part.file_data is not None:
-                    has_attachments = True
-                    break
-            if has_attachments:
-                break
-
-        logging.info(f"GoogleLLM: Starting stream generation. Model: {model}, Messages: {json.dumps(messages, default=str)}, Has attachments: {has_attachments}")
-
         response = client.models.generate_content_stream(
             model=model,
             contents=messages,
             config=config,
         )
-
-
         for chunk in response:
             if hasattr(chunk, "candidates") and chunk.candidates:
                 for candidate in chunk.candidates:

application/llm/handlers/base.py

@@ -1,335 +0,0 @@
-import logging
-from abc import ABC, abstractmethod
-from dataclasses import dataclass
-from typing import Any, Dict, Generator, List, Optional, Union
-
-from application.logging import build_stack_data
-
-logger = logging.getLogger(__name__)
-
-
-@dataclass
-class ToolCall:
-    """Represents a tool/function call from the LLM."""
-
-    id: str
-    name: str
-    arguments: Union[str, Dict]
-    index: Optional[int] = None
-
-    @classmethod
-    def from_dict(cls, data: Dict) -> "ToolCall":
-        """Create ToolCall from dictionary."""
-        return cls(
-            id=data.get("id", ""),
-            name=data.get("name", ""),
-            arguments=data.get("arguments", {}),
-            index=data.get("index"),
-        )
-
-
-@dataclass
-class LLMResponse:
-    """Represents a response from the LLM."""
-
-    content: str
-    tool_calls: List[ToolCall]
-    finish_reason: str
-    raw_response: Any
-
-    @property
-    def requires_tool_call(self) -> bool:
-        """Check if the response requires tool calls."""
-        return bool(self.tool_calls) and self.finish_reason == "tool_calls"
-
-
-class LLMHandler(ABC):
-    """Abstract base class for LLM handlers."""
-
-    def __init__(self):
-        self.llm_calls = []
-        self.tool_calls = []
-
-    @abstractmethod
-    def parse_response(self, response: Any) -> LLMResponse:
-        """Parse raw LLM response into standardized format."""
-        pass
-
-    @abstractmethod
-    def create_tool_message(self, tool_call: ToolCall, result: Any) -> Dict:
-        """Create a tool result message for the conversation history."""
-        pass
-
-    @abstractmethod
-    def _iterate_stream(self, response: Any) -> Generator:
-        """Iterate through streaming response chunks."""
-        pass
-
-    def process_message_flow(
-        self,
-        agent,
-        initial_response,
-        tools_dict: Dict,
-        messages: List[Dict],
-        attachments: Optional[List] = None,
-        stream: bool = False,
-    ) -> Union[str, Generator]:
-        """
-        Main orchestration method for processing LLM message flow.
-
-        Args:
-            agent: The agent instance
-            initial_response: Initial LLM response
-            tools_dict: Dictionary of available tools
-            messages: Conversation history
-            attachments: Optional attachments
-            stream: Whether to use streaming
-
-        Returns:
-            Final response or generator for streaming
-        """
-        messages = self.prepare_messages(agent, messages, attachments)
-
-        if stream:
-            return self.handle_streaming(agent, initial_response, tools_dict, messages)
-        else:
-            return self.handle_non_streaming(
-                agent, initial_response, tools_dict, messages
-            )
-
-    def prepare_messages(
-        self, agent, messages: List[Dict], attachments: Optional[List] = None
-    ) -> List[Dict]:
-        """
-        Prepare messages with attachments and provider-specific formatting.
-
-        Args:
-            agent: The agent instance
-            messages: Original messages
-            attachments: List of attachments
-
-        Returns:
-            Prepared messages list
-        """
-        if not attachments:
-            return messages
-        logger.info(f"Preparing messages with {len(attachments)} attachments")
-        supported_types = agent.llm.get_supported_attachment_types()
-
-        supported_attachments = [
-            a for a in attachments if a.get("mime_type") in supported_types
-        ]
-        unsupported_attachments = [
-            a for a in attachments if a.get("mime_type") not in supported_types
-        ]
-
-        # Process supported attachments with the LLM's custom method
-
-        if supported_attachments:
-            logger.info(
-                f"Processing {len(supported_attachments)} supported attachments"
-            )
-            messages = agent.llm.prepare_messages_with_attachments(
-                messages, supported_attachments
-            )
-        # Process unsupported attachments with default method
-
-        if unsupported_attachments:
-            logger.info(
-                f"Processing {len(unsupported_attachments)} unsupported attachments"
-            )
-            messages = self._append_unsupported_attachments(
-                messages, unsupported_attachments
-            )
-        return messages
-
-    def _append_unsupported_attachments(
-        self, messages: List[Dict], attachments: List[Dict]
-    ) -> List[Dict]:
-        """
-        Default method to append unsupported attachment content to system prompt.
-
-        Args:
-            messages: Current messages
-            attachments: List of unsupported attachments
-
-        Returns:
-            Updated messages list
-        """
-        prepared_messages = messages.copy()
-        attachment_texts = []
-
-        for attachment in attachments:
-            logger.info(f"Adding attachment {attachment.get('id')} to context")
-            if "content" in attachment:
-                attachment_texts.append(
-                    f"Attached file content:\n\n{attachment['content']}"
-                )
-        if attachment_texts:
-            combined_text = "\n\n".join(attachment_texts)
-
-            system_msg = next(
-                (msg for msg in prepared_messages if msg.get("role") == "system"),
-                {"role": "system", "content": ""},
-            )
-
-            if system_msg not in prepared_messages:
-                prepared_messages.insert(0, system_msg)
-            system_msg["content"] += f"\n\n{combined_text}"
-        return prepared_messages
-
-    def handle_tool_calls(
-        self, agent, tool_calls: List[ToolCall], tools_dict: Dict, messages: List[Dict]
-    ) -> Generator:
-        """
-        Execute tool calls and update conversation history.
-
-        Args:
-            agent: The agent instance
-            tool_calls: List of tool calls to execute
-            tools_dict: Available tools dictionary
-            messages: Current conversation history
-
-        Returns:
-            Updated messages list
-        """
-        updated_messages = messages.copy()
-
-        for call in tool_calls:
-            try:
-                self.tool_calls.append(call)
-                tool_executor_gen = agent._execute_tool_action(tools_dict, call)
-                while True:
-                    try:
-                        yield next(tool_executor_gen)
-                    except StopIteration as e:
-                        tool_response, call_id = e.value
-                        break
-
-                updated_messages.append(
-                    {
-                        "role": "assistant",
-                        "content": [
-                            {
-                                "function_call": {
-                                    "name": call.name,
-                                    "args": call.arguments,
-                                    "call_id": call_id,
-                                }
-                            }
-                        ],
-                    }
-                )
-
-                updated_messages.append(self.create_tool_message(call, tool_response))
-
-            except Exception as e:
-                logger.error(f"Error executing tool: {str(e)}", exc_info=True)
-                updated_messages.append(
-                    {
-                        "role": "tool",
-                        "content": f"Error executing tool: {str(e)}",
-                        "tool_call_id": call.id,
-                    }
-                )
-
-        return updated_messages
-
-    def handle_non_streaming(
-        self, agent, response: Any, tools_dict: Dict, messages: List[Dict]
-    ) -> Generator:
-        """
-        Handle non-streaming response flow.
-
-        Args:
-            agent: The agent instance
-            response: Current LLM response
-            tools_dict: Available tools dictionary
-            messages: Conversation history
-
-        Returns:
-            Final response after processing all tool calls
-        """
-        parsed = self.parse_response(response)
-        self.llm_calls.append(build_stack_data(agent.llm))
-
-        while parsed.requires_tool_call:
-            tool_handler_gen = self.handle_tool_calls(
-                agent, parsed.tool_calls, tools_dict, messages
-            )
-            while True:
-                try:
-                    yield next(tool_handler_gen)
-                except StopIteration as e:
-                    messages = e.value
-                    break
-
-            response = agent.llm.gen(
-                model=agent.gpt_model, messages=messages, tools=agent.tools
-            )
-            parsed = self.parse_response(response)
-            self.llm_calls.append(build_stack_data(agent.llm))
-
-        return parsed.content
-
-    def handle_streaming(
-        self, agent, response: Any, tools_dict: Dict, messages: List[Dict]
-    ) -> Generator:
-        """
-        Handle streaming response flow.
-
-        Args:
-            agent: The agent instance
-            response: Current LLM response
-            tools_dict: Available tools dictionary
-            messages: Conversation history
-
-        Yields:
-            Streaming response chunks
-        """
-        buffer = ""
-        tool_calls = {}
-
-        for chunk in self._iterate_stream(response):
-            if isinstance(chunk, str):
-                yield chunk
-                continue
-            parsed = self.parse_response(chunk)
-
-            if parsed.tool_calls:
-                for call in parsed.tool_calls:
-                    if call.index not in tool_calls:
-                        tool_calls[call.index] = call
-                    else:
-                        existing = tool_calls[call.index]
-                        if call.id:
-                            existing.id = call.id
-                        if call.name:
-                            existing.name = call.name
-                        if call.arguments:
-                            existing.arguments += call.arguments
-            if parsed.finish_reason == "tool_calls":
-                tool_handler_gen = self.handle_tool_calls(
-                    agent, list(tool_calls.values()), tools_dict, messages
-                )
-                while True:
-                    try:
-                        yield next(tool_handler_gen)
-                    except StopIteration as e:
-                        messages = e.value
-                        break
-                tool_calls = {}
-
-                response = agent.llm.gen_stream(
-                    model=agent.gpt_model, messages=messages, tools=agent.tools
-                )
-                self.llm_calls.append(build_stack_data(agent.llm))
-
-                yield from self.handle_streaming(agent, response, tools_dict, messages)
-                return
-            if parsed.content:
-                buffer += parsed.content
-                yield buffer
-                buffer = ""
-            if parsed.finish_reason == "stop":
-                return

application/llm/handlers/google.py

@@ -1,78 +0,0 @@
-import uuid
-from typing import Any, Dict, Generator
-
-from application.llm.handlers.base import LLMHandler, LLMResponse, ToolCall
-
-
-class GoogleLLMHandler(LLMHandler):
-    """Handler for Google's GenAI API."""
-
-    def parse_response(self, response: Any) -> LLMResponse:
-        """Parse Google response into standardized format."""
-
-        if isinstance(response, str):
-            return LLMResponse(
-                content=response,
-                tool_calls=[],
-                finish_reason="stop",
-                raw_response=response,
-            )
-
-        if hasattr(response, "candidates"):
-            parts = response.candidates[0].content.parts if response.candidates else []
-            tool_calls = [
-                ToolCall(
-                    id=str(uuid.uuid4()),
-                    name=part.function_call.name,
-                    arguments=part.function_call.args,
-                )
-                for part in parts
-                if hasattr(part, "function_call") and part.function_call is not None
-            ]
-
-            content = " ".join(
-                part.text
-                for part in parts
-                if hasattr(part, "text") and part.text is not None
-            )
-            return LLMResponse(
-                content=content,
-                tool_calls=tool_calls,
-                finish_reason="tool_calls" if tool_calls else "stop",
-                raw_response=response,
-            )
-
-        else:
-            tool_calls = []
-            if hasattr(response, "function_call"):
-                tool_calls.append(
-                    ToolCall(
-                        id=str(uuid.uuid4()),
-                        name=response.function_call.name,
-                        arguments=response.function_call.args,
-                    )
-                )
-            return LLMResponse(
-                content=response.text if hasattr(response, "text") else "",
-                tool_calls=tool_calls,
-                finish_reason="tool_calls" if tool_calls else "stop",
-                raw_response=response,
-            )
-
-    def create_tool_message(self, tool_call: ToolCall, result: Any) -> Dict:
-        """Create Google-style tool message."""
-        from google.genai import types
-
-        return {
-            "role": "tool",
-            "content": [
-                types.Part.from_function_response(
-                    name=tool_call.name, response={"result": result}
-                ).to_json_dict()
-            ],
-        }
-
-    def _iterate_stream(self, response: Any) -> Generator:
-        """Iterate through Google streaming response."""
-        for chunk in response:
-            yield chunk

application/llm/handlers/handler_creator.py

@@ -1,18 +0,0 @@
-from application.llm.handlers.base import LLMHandler
-from application.llm.handlers.google import GoogleLLMHandler
-from application.llm.handlers.openai import OpenAILLMHandler
-
-
-class LLMHandlerCreator:
-    handlers = {
-        "openai": OpenAILLMHandler,
-        "google": GoogleLLMHandler,
-        "default": OpenAILLMHandler,
-    }
-
-    @classmethod
-    def create_handler(cls, llm_type: str, *args, **kwargs) -> LLMHandler:
-        handler_class = cls.handlers.get(llm_type.lower())
-        if not handler_class:
-            handler_class = OpenAILLMHandler
-        return handler_class(*args, **kwargs)

application/llm/handlers/openai.py

@@ -1,57 +0,0 @@
-from typing import Any, Dict, Generator
-
-from application.llm.handlers.base import LLMHandler, LLMResponse, ToolCall
-
-
-class OpenAILLMHandler(LLMHandler):
-    """Handler for OpenAI API."""
-
-    def parse_response(self, response: Any) -> LLMResponse:
-        """Parse OpenAI response into standardized format."""
-        if isinstance(response, str):
-            return LLMResponse(
-                content=response,
-                tool_calls=[],
-                finish_reason="stop",
-                raw_response=response,
-            )
-
-        message = getattr(response, "message", None) or getattr(response, "delta", None)
-
-        tool_calls = []
-        if hasattr(message, "tool_calls"):
-            tool_calls = [
-                ToolCall(
-                    id=getattr(tc, "id", ""),
-                    name=getattr(tc.function, "name", ""),
-                    arguments=getattr(tc.function, "arguments", ""),
-                    index=getattr(tc, "index", None),
-                )
-                for tc in message.tool_calls or []
-            ]
-        return LLMResponse(
-            content=getattr(message, "content", ""),
-            tool_calls=tool_calls,
-            finish_reason=getattr(response, "finish_reason", ""),
-            raw_response=response,
-        )
-
-    def create_tool_message(self, tool_call: ToolCall, result: Any) -> Dict:
-        """Create OpenAI-style tool message."""
-        return {
-            "role": "tool",
-            "content": [
-                {
-                    "function_response": {
-                        "name": tool_call.name,
-                        "response": {"result": result},
-                        "call_id": tool_call.id,
-                    }
-                }
-            ],
-        }
-
-    def _iterate_stream(self, response: Any) -> Generator:
-        """Iterate through OpenAI streaming response."""
-        for chunk in response:
-            yield chunk

application/llm/llama_cpp.py

@@ -2,7 +2,6 @@ from application.llm.base import BaseLLM
 from application.core.settings import settings
 import threading
 
-
 class LlamaSingleton:
     _instances = {}
     _lock = threading.Lock()  # Add a lock for thread synchronization
@@ -30,7 +29,7 @@ class LlamaCpp(BaseLLM):
         self,
         api_key=None,
         user_api_key=None,
-        llm_name=settings.LLM_PATH,
+        llm_name=settings.MODEL_PATH,
         *args,
         **kwargs,
     ):
@@ -43,18 +42,14 @@ class LlamaCpp(BaseLLM):
         context = messages[0]["content"]
         user_question = messages[-1]["content"]
         prompt = f"### Instruction \n {user_question} \n ### Context \n {context} \n ### Answer \n"
-        result = LlamaSingleton.query_model(
-            self.llama, prompt, max_tokens=150, echo=False
-        )
+        result = LlamaSingleton.query_model(self.llama, prompt, max_tokens=150, echo=False)
         return result["choices"][0]["text"].split("### Answer \n")[-1]
 
     def _raw_gen_stream(self, baseself, model, messages, stream=True, **kwargs):
         context = messages[0]["content"]
         user_question = messages[-1]["content"]
         prompt = f"### Instruction \n {user_question} \n ### Context \n {context} \n ### Answer \n"
-        result = LlamaSingleton.query_model(
-            self.llama, prompt, max_tokens=150, echo=False, stream=stream
-        )
+        result = LlamaSingleton.query_model(self.llama, prompt, max_tokens=150, echo=False, stream=stream)
         for item in result:
             for choice in item["choices"]:
-                yield choice["text"]
+                yield choice["text"]

application/llm/openai.py

@@ -1,10 +1,7 @@
 import json
-import base64
-import logging
 
 from application.core.settings import settings
 from application.llm.base import BaseLLM
-from application.storage.storage_creator import StorageCreator
 
 
 class OpenAILLM(BaseLLM):
@@ -13,14 +10,12 @@ class OpenAILLM(BaseLLM):
         from openai import OpenAI
 
         super().__init__(*args, **kwargs)
-        if isinstance(settings.OPENAI_BASE_URL, str) and settings.OPENAI_BASE_URL.strip():
+        if settings.OPENAI_BASE_URL:
             self.client = OpenAI(api_key=api_key, base_url=settings.OPENAI_BASE_URL)
         else:
-            DEFAULT_OPENAI_API_BASE = "https://api.openai.com/v1"
-            self.client = OpenAI(api_key=api_key, base_url=DEFAULT_OPENAI_API_BASE)
+            self.client = OpenAI(api_key=api_key)
         self.api_key = api_key
         self.user_api_key = user_api_key
-        self.storage = StorageCreator.get_storage()
 
     def _clean_messages_openai(self, messages):
         cleaned_messages = []
@@ -70,17 +65,6 @@ class OpenAILLM(BaseLLM):
                                     ),
                                 }
                             )
-                        elif isinstance(item, dict):
-                            content_parts = []
-                            if "text" in item:
-                                content_parts.append({"type": "text", "text": item["text"]})
-                            elif "type" in item and item["type"] == "text" and "text" in item:
-                                content_parts.append(item)
-                            elif "type" in item and item["type"] == "file" and "file" in item:
-                                content_parts.append(item)
-                            elif "type" in item and item["type"] == "image_url" and "image_url" in item:
-                                content_parts.append(item)
-                            cleaned_messages.append({"role": role, "content": content_parts})
                         else:
                             raise ValueError(
                                 f"Unexpected content dictionary format: {item}"
@@ -149,167 +133,11 @@ class OpenAILLM(BaseLLM):
     def _supports_tools(self):
         return True
 
-    def get_supported_attachment_types(self):
-        """
-        Return a list of MIME types supported by OpenAI for file uploads.
-
-        Returns:
-            list: List of supported MIME types
-        """
-        return [
-            'application/pdf',
-            'image/png',
-            'image/jpeg',
-            'image/jpg',
-            'image/webp',
-            'image/gif'
-        ]
-
-    def prepare_messages_with_attachments(self, messages, attachments=None):
-        """
-        Process attachments using OpenAI's file API for more efficient handling.
-
-        Args:
-            messages (list): List of message dictionaries.
-            attachments (list): List of attachment dictionaries with content and metadata.
-
-        Returns:
-            list: Messages formatted with file references for OpenAI API.
-        """
-        if not attachments:
-            return messages
-
-        prepared_messages = messages.copy()
-
-        # Find the user message to attach file_id to the last one
-        user_message_index = None
-        for i in range(len(prepared_messages) - 1, -1, -1):
-            if prepared_messages[i].get("role") == "user":
-                user_message_index = i
-                break
-
-        if user_message_index is None:
-            user_message = {"role": "user", "content": []}
-            prepared_messages.append(user_message)
-            user_message_index = len(prepared_messages) - 1
-
-        if isinstance(prepared_messages[user_message_index].get("content"), str):
-            text_content = prepared_messages[user_message_index]["content"]
-            prepared_messages[user_message_index]["content"] = [
-                {"type": "text", "text": text_content}
-            ]
-        elif not isinstance(prepared_messages[user_message_index].get("content"), list):
-            prepared_messages[user_message_index]["content"] = []
-
-        for attachment in attachments:
-            mime_type = attachment.get('mime_type')
-
-            if mime_type and mime_type.startswith('image/'):
-                try:
-                    base64_image = self._get_base64_image(attachment)
-                    prepared_messages[user_message_index]["content"].append({
-                        "type": "image_url",
-                        "image_url": {
-                            "url": f"data:{mime_type};base64,{base64_image}"
-                        }
-                    })
-                except Exception as e:
-                    logging.error(f"Error processing image attachment: {e}", exc_info=True)
-                    if 'content' in attachment:
-                        prepared_messages[user_message_index]["content"].append({
-                            "type": "text",
-                            "text": f"[Image could not be processed: {attachment.get('path', 'unknown')}]"
-                        })
-            # Handle PDFs using the file API
-            elif mime_type == 'application/pdf':
-                try:
-                    file_id = self._upload_file_to_openai(attachment)
-                    prepared_messages[user_message_index]["content"].append({
-                        "type": "file",
-                        "file": {"file_id": file_id}
-                    })
-                except Exception as e:
-                    logging.error(f"Error uploading PDF to OpenAI: {e}", exc_info=True)
-                    if 'content' in attachment:
-                        prepared_messages[user_message_index]["content"].append({
-                            "type": "text",
-                            "text": f"File content:\n\n{attachment['content']}"
-                        })
-
-        return prepared_messages
-
-    def _get_base64_image(self, attachment):
-        """
-        Convert an image file to base64 encoding.
-
-        Args:
-            attachment (dict): Attachment dictionary with path and metadata.
-
-        Returns:
-            str: Base64-encoded image data.
-        """
-        file_path = attachment.get('path')
-        if not file_path:
-            raise ValueError("No file path provided in attachment")
-
-        try:
-            with self.storage.get_file(file_path) as image_file:
-                return base64.b64encode(image_file.read()).decode('utf-8')
-        except FileNotFoundError:
-            raise FileNotFoundError(f"File not found: {file_path}")
-
-    def _upload_file_to_openai(self, attachment):
-        """
-        Upload a file to OpenAI and return the file_id.
-
-        Args:
-            attachment (dict): Attachment dictionary with path and metadata.
-                Expected keys:
-                - path: Path to the file
-                - id: Optional MongoDB ID for caching
-
-        Returns:
-            str: OpenAI file_id for the uploaded file.
-        """
-        import logging
-
-        if 'openai_file_id' in attachment:
-            return attachment['openai_file_id']
-
-        file_path = attachment.get('path')
-
-        if not self.storage.file_exists(file_path):
-            raise FileNotFoundError(f"File not found: {file_path}")
-
-        try:
-            file_id = self.storage.process_file(
-                file_path,
-                lambda local_path, **kwargs: self.client.files.create(
-                    file=open(local_path, 'rb'),
-                    purpose="assistants"
-                ).id
-            )
-
-            from application.core.mongo_db import MongoDB
-            mongo = MongoDB.get_client()
-            db = mongo[settings.MONGO_DB_NAME]
-            attachments_collection = db["attachments"]
-            if '_id' in attachment:
-                attachments_collection.update_one(
-                    {"_id": attachment['_id']},
-                    {"$set": {"openai_file_id": file_id}}
-                )
-
-            return file_id
-        except Exception as e:
-            logging.error(f"Error uploading file to OpenAI: {e}", exc_info=True)
-            raise
-
 
 class AzureOpenAILLM(OpenAILLM):
 
     def __init__(
-        self, api_key, user_api_key, *args, **kwargs
+        self, api_key, user_api_key, *args, **kwargs 
     ):
 
         super().__init__(api_key)

application/logging.py

@@ -7,7 +7,6 @@ import uuid
 from typing import Any, Callable, Dict, Generator, List
 
 from application.core.mongo_db import MongoDB
-from application.core.settings import settings
 
 logging.basicConfig(
     level=logging.INFO, format="%(asctime)s - %(levelname)s - %(message)s"
@@ -30,8 +29,6 @@ def build_stack_data(
     exclude_attributes: List[str] = None,
     custom_data: Dict = None,
 ) -> Dict:
-    if obj is None:
-        raise ValueError("The 'obj' parameter cannot be None")
     data = {}
     if include_attributes is None:
         include_attributes = []
@@ -59,8 +56,8 @@ def build_stack_data(
                         data[attr_name] = [str(item) for item in attr_value]
                 elif isinstance(attr_value, dict):
                     data[attr_name] = {k: str(v) for k, v in attr_value.items()}
-        except AttributeError as e:
-            logging.warning(f"AttributeError while accessing {attr_name}: {e}")
+                else:
+                    data[attr_name] = str(attr_value)
         except AttributeError:
             pass
     if custom_data:
@@ -134,7 +131,7 @@ def _log_to_mongodb(
 ) -> None:
     try:
         mongo = MongoDB.get_client()
-        db = mongo[settings.MONGO_DB_NAME]
+        db = mongo["docsgpt"]
         user_logs_collection = db["stack_logs"]
 
         log_entry = {
@@ -151,4 +148,4 @@ def _log_to_mongodb(
         logging.debug(f"Logged activity to MongoDB: {activity_id}")
 
     except Exception as e:
-        logging.error(f"Failed to log to MongoDB: {e}", exc_info=True)
+        logging.error(f"Failed to log to MongoDB: {e}")

application/parser/embedding_pipeline.py

@@ -19,7 +19,7 @@ def add_text_to_store_with_retry(store, doc, source_id):
         doc.metadata["source_id"] = str(source_id)
         store.add_texts([doc.page_content], metadatas=[doc.metadata])
     except Exception as e:
-        logging.error(f"Failed to add document with retry: {e}", exc_info=True)
+        logging.error(f"Failed to add document with retry: {e}")
         raise
 
 
@@ -75,7 +75,7 @@ def embed_and_store_documents(docs, folder_name, source_id, task_status):
             # Add document to vector store
             add_text_to_store_with_retry(store, doc, source_id)
         except Exception as e:
-            logging.error(f"Error embedding document {idx}: {e}", exc_info=True)
+            logging.error(f"Error embedding document {idx}: {e}")
             logging.info(f"Saving progress at document {idx} out of {total_docs}")
             store.save_local(folder_name)
             break

application/parser/file/bulk.py

@@ -158,7 +158,7 @@ class SimpleDirectoryReader(BaseReader):
                     data = f.read()
             # Prepare metadata for this file
             if self.file_metadata is not None:
-                file_metadata = self.file_metadata(input_file.name)
+                file_metadata = self.file_metadata(str(input_file))
             else:
                 # Provide a default empty metadata
                 file_metadata = {'title': '', 'store': ''}

application/parser/file/tabular_parser.py

@@ -73,13 +73,7 @@ class PandasCSVParser(BaseParser):
             for more information.
             Set to empty dict by default, this means pandas will try to figure
             out the separators, table head, etc. on its own.
-            
-        header_period (int): Controls how headers are included in output:
-            - 0: Headers only at the beginning
-            - 1: Headers in every row
-            - N > 1: Headers every N rows
-            
-        header_prefix (str): Prefix for header rows. Default is "HEADERS: ".
+
     """
 
     def __init__(
@@ -89,8 +83,6 @@ class PandasCSVParser(BaseParser):
             col_joiner: str = ", ",
             row_joiner: str = "\n",
             pandas_config: dict = {},
-            header_period: int = 20,
-            header_prefix: str = "HEADERS: ",
             **kwargs: Any
     ) -> None:
         """Init params."""
@@ -99,8 +91,6 @@ class PandasCSVParser(BaseParser):
         self._col_joiner = col_joiner
         self._row_joiner = row_joiner
         self._pandas_config = pandas_config
-        self._header_period = header_period
-        self._header_prefix = header_prefix
 
     def _init_parser(self) -> Dict:
         """Init parser."""
@@ -114,26 +104,15 @@ class PandasCSVParser(BaseParser):
             raise ValueError("pandas module is required to read CSV files.")
 
         df = pd.read_csv(file, **self._pandas_config)
-        headers = df.columns.tolist()
-        header_row = f"{self._header_prefix}{self._col_joiner.join(headers)}"
 
-        if not self._concat_rows:
-            return df.apply(
-                lambda row: (self._col_joiner).join(row.astype(str).tolist()), axis=1
-            ).tolist()
-        
-        text_list = []
-        if self._header_period != 1:
-            text_list.append(header_row)
-        
-        for i, row in df.iterrows():
-            if (self._header_period > 1 and i > 0 and i % self._header_period == 0):
-                text_list.append(header_row)
-            text_list.append(self._col_joiner.join(row.astype(str).tolist()))
-            if self._header_period == 1 and i < len(df) - 1:
-                text_list.append(header_row)
+        text_list = df.apply(
+            lambda row: (self._col_joiner).join(row.astype(str).tolist()), axis=1
+        ).tolist()
 
-        return self._row_joiner.join(text_list)
+        if self._concat_rows:
+            return (self._row_joiner).join(text_list)
+        else:
+            return text_list
 
 
 class ExcelParser(BaseParser):
@@ -159,13 +138,7 @@ class ExcelParser(BaseParser):
             for more information.
             Set to empty dict by default, this means pandas will try to figure
             out the table structure on its own.
-            
-        header_period (int): Controls how headers are included in output:
-            - 0: Headers only at the beginning (default)
-            - 1: Headers in every row
-            - N > 1: Headers every N rows
-            
-        header_prefix (str): Prefix for header rows. Default is "HEADERS: ".
+
     """
 
     def __init__(
@@ -175,8 +148,6 @@ class ExcelParser(BaseParser):
             col_joiner: str = ", ",
             row_joiner: str = "\n",
             pandas_config: dict = {},
-            header_period: int = 20,
-            header_prefix: str = "HEADERS: ",
             **kwargs: Any
     ) -> None:
         """Init params."""
@@ -185,8 +156,6 @@ class ExcelParser(BaseParser):
         self._col_joiner = col_joiner
         self._row_joiner = row_joiner
         self._pandas_config = pandas_config
-        self._header_period = header_period
-        self._header_prefix = header_prefix
 
     def _init_parser(self) -> Dict:
         """Init parser."""
@@ -200,22 +169,12 @@ class ExcelParser(BaseParser):
             raise ValueError("pandas module is required to read Excel files.")
 
         df = pd.read_excel(file, **self._pandas_config)
-        headers = df.columns.tolist()
-        header_row = f"{self._header_prefix}{self._col_joiner.join(headers)}"
-        
-        if not self._concat_rows:
-            return df.apply(
-                lambda row: (self._col_joiner).join(row.astype(str).tolist()), axis=1
-            ).tolist()
-        
-        text_list = []
-        if self._header_period != 1:
-            text_list.append(header_row)
 
-        for i, row in df.iterrows():
-            if (self._header_period > 1 and i > 0 and i % self._header_period == 0):
-                text_list.append(header_row)
-            text_list.append(self._col_joiner.join(row.astype(str).tolist()))
-            if self._header_period == 1 and i < len(df) - 1:
-                text_list.append(header_row)
-        return self._row_joiner.join(text_list)
+        text_list = df.apply(
+            lambda row: (self._col_joiner).join(row.astype(str).tolist()), axis=1
+        ).tolist()
+
+        if self._concat_rows:
+            return (self._row_joiner).join(text_list)
+        else:
+            return text_list

application/parser/remote/crawler_loader.py

@@ -1,4 +1,3 @@
-import logging
 import requests
 from urllib.parse import urlparse, urljoin
 from bs4 import BeautifulSoup
@@ -43,7 +42,7 @@ class CrawlerLoader(BaseRemote):
                         )
                     )
             except Exception as e:
-                logging.error(f"Error processing URL {current_url}: {e}", exc_info=True)
+                print(f"Error processing URL {current_url}: {e}")
                 continue
 
             # Parse the HTML content to extract all links
@@ -62,4 +61,4 @@ class CrawlerLoader(BaseRemote):
             if self.limit is not None and len(visited_urls) >= self.limit:
                 break
 
-        return loaded_content
+        return loaded_content

application/parser/remote/sitemap_loader.py

@@ -1,4 +1,3 @@
-import logging
 import requests
 import re  # Import regular expression library
 import xml.etree.ElementTree as ET
@@ -33,7 +32,7 @@ class SitemapLoader(BaseRemote):
                 documents.extend(loader.load())
                 processed_urls += 1  # Increment the counter after processing each URL
             except Exception as e:
-                logging.error(f"Error processing URL {url}: {e}", exc_info=True)
+                print(f"Error processing URL {url}: {e}")
                 continue
 
         return documents

application/parser/remote/web_loader.py

@@ -1,4 +1,3 @@
-import logging
 from application.parser.remote.base import BaseRemote
 from application.parser.schema.base import Document
 from langchain_community.document_loaders import WebBaseLoader
@@ -40,6 +39,6 @@ class WebLoader(BaseRemote):
                         )
                     )
             except Exception as e:
-                logging.error(f"Error processing URL {url}: {e}", exc_info=True)
+                print(f"Error processing URL {url}: {e}")
                 continue
-        return documents
+        return documents

application/prompts/chat_combine_creative.txt

@@ -1,15 +1,9 @@
-You are a helpful AI assistant, DocsGPT. You are proactive and helpful. Try to use tools, if they are available to you, 
-be proactive and fill in missing information.
-Users can Upload documents for your context as attachments or sources via UI using the Conversation input box.
-If appropriate, your answers can include code examples, formatted as follows:
+You are a DocsGPT, friendly and helpful AI assistant by Arc53 that provides help with documents. You give thorough answers with code examples if possible.
+Use the following pieces of context to help answer the users question. If its not relevant to the question, provide friendly responses.
+You have access to chat history, and can use it to help answer the question.
+When using code examples, use the following format:
 ```(language)
 (code)
 ```
-Users are also able to see charts and diagrams if you use them with valid mermaid syntax in your responses.
-Try to respond with mermaid charts if visualization helps with users queries.
-You effectively utilize chat history, ensuring relevant and tailored responses. 
-Try to use additional provided context if it's available, otherwise use your knowledge and tool capabilities.
-Allow yourself to be very creative and use your imagination.
 ----------------
-Possible additional context from uploaded sources:
 {summaries}

application/prompts/chat_combine_default.txt

@@ -1,14 +1,9 @@
-You are a helpful AI assistant, DocsGPT. You are proactive and helpful. Try to use tools, if they are available to you, 
-be proactive and fill in missing information.
-Users can Upload documents for your context as attachments or sources via UI using the Conversation input box.
+You are a helpful AI assistant, DocsGPT, specializing in document assistance, designed to offer detailed and informative responses. 
 If appropriate, your answers can include code examples, formatted as follows:
 ```(language)
 (code)
 ```
-Users are also able to see charts and diagrams if you use them with valid mermaid syntax in your responses.
-Try to respond with mermaid charts if visualization helps with users queries.
 You effectively utilize chat history, ensuring relevant and tailored responses. 
-Try to use additional provided context if it's available, otherwise use your knowledge and tool capabilities.
+If a question doesn't align with your context, you provide friendly and helpful replies.
 ----------------
-Possible additional context from uploaded sources:
 {summaries}

application/prompts/chat_combine_strict.txt

@@ -1,17 +1,13 @@
-You are a helpful AI assistant, DocsGPT. You are proactive and helpful. Try to use tools, if they are available to you, 
-be proactive and fill in missing information.
-Users can Upload documents for your context as attachments or sources via UI using the Conversation input box.
-If appropriate, your answers can include code examples, formatted as follows:
+You are an AI Assistant, DocsGPT, adept at offering document assistance. 
+Your expertise lies in providing answer on top of provided context. 
+You can leverage the chat history if needed.
+Answer the question based on the context below. 
+Keep the answer concise. Respond "Irrelevant context" if not sure about the answer.
+If question is not related to the context, respond "Irrelevant context".
+When using code examples, use the following format:
 ```(language)
 (code)
 ```
-Users are also able to see charts and diagrams if you use them with valid mermaid syntax in your responses.
-Try to respond with mermaid charts if visualization helps with users queries.
-You effectively utilize chat history, ensuring relevant and tailored responses. 
-Use context provided below or use available tools tool capabilities to answer user queries.
-If you dont have enough information from the context or tools, answer "I don't know" or "I don't have enough information".
-Never make up information or provide false information!
-Allow yourself to be very creative and use your imagination.
-----------------
-Context from uploaded sources:
-{summaries}
+ ----------------
+ Context:
+ {summaries}

application/prompts/react_final_prompt.txt

@@ -1,3 +0,0 @@
-Query: {query}
-Observations: {observations}
-Now, using the insights from the observations, formulate a well-structured and precise final answer.

application/prompts/react_planning_prompt.txt

@@ -1,13 +0,0 @@
-You are an AI assistant and talk like you're thinking out loud. Given the following query, outline a concise thought process that includes key steps and considerations necessary for effective analysis and response. Avoid pointwise formatting. The goal is to break down the query into manageable components without excessive detail, focusing on clarity and logical progression.
-
-Include the following elements in your thought and execution process:
-1. Identify the main objective of the query.
-2. Determine any relevant context or background information needed to understand the query.
-3. List potential approaches or methods to address the query.
-4. Highlight any critical factors or constraints that may influence the outcome.
-5. Plan with available tools to help you with the analysis but dont execute them. Tools will be executed by another AI.
-
-Query: {query}  
-Summaries: {summaries}
-Prompt: {prompt}
-Observations(potentially previous tool calls): {observations}

application/requirements.txt

@@ -1,20 +1,25 @@
 anthropic==0.49.0
-boto3==1.38.18
-beautifulsoup4==4.13.4
+boto3==1.35.97
+beautifulsoup4==4.12.3
 celery==5.4.0
 dataclasses-json==0.6.7
 docx2txt==0.8
 duckduckgo-search==7.5.2
 ebooklib==0.18
+elastic-transport==8.17.0
+elasticsearch==8.17.1
 escodegen==1.0.11
 esprima==4.0.1
 esutils==1.0.1
-Flask==3.1.1
+Flask==3.1.0
 faiss-cpu==1.9.0.post1
 flask-restx==1.3.0
+gevent==24.11.1
 google-genai==1.3.0
+google-generativeai==0.8.3
 gTTS==2.5.4
 gunicorn==23.0.0
+html2text==2024.2.26
 javalang==0.13.0
 jinja2==3.1.6
 jiter==0.8.2
@@ -22,56 +27,63 @@ jmespath==1.0.1
 joblib==1.4.2
 jsonpatch==1.33
 jsonpointer==3.0.0
+jsonschema==4.23.0
+jsonschema-spec==0.2.4
+jsonschema-specifications==2023.7.1
 kombu==5.4.2
 langchain==0.3.20
 langchain-community==0.3.19
-langchain-core==0.3.59
-langchain-openai==0.3.16
-langchain-text-splitters==0.3.8
-langsmith==0.3.42
+langchain-core==0.3.45
+langchain-openai==0.3.8
+langchain-text-splitters==0.3.6
+langsmith==0.3.15
 lazy-object-proxy==1.10.0
 lxml==5.3.1
 markupsafe==3.0.2
 marshmallow==3.26.1
 mpmath==1.3.0
-multidict==6.4.3
+multidict==6.1.0
 mypy-extensions==1.0.0
 networkx==3.4.2
 numpy==2.2.1
-openai==1.78.1
-openapi3-parser==1.1.21
+openai==1.66.3
+openapi-schema-validator==0.6.3
+openapi-spec-validator==0.6.0
+openapi3-parser==1.1.19
 orjson==3.10.14
-packaging==24.2
+packaging==24.1
 pandas==2.2.3
 openpyxl==3.1.5
 pathable==0.4.4
 pillow==11.1.0
-portalocker>=2.7.0,<3.0.0
+portalocker==2.10.1
 prance==23.6.21.0
-prompt-toolkit==3.0.51
+primp==0.14.0
+prompt-toolkit==3.0.50
 protobuf==5.29.3
 psycopg2-binary==2.9.10
 py==1.11.0
 pydantic==2.10.6
 pydantic-core==2.27.2
 pydantic-settings==2.7.1
-pymongo==4.11.3
-pypdf==5.5.0
+pymongo==4.10.1
+pypdf==5.2.0
 python-dateutil==2.9.0.post0
 python-dotenv==1.0.1
 python-jose==3.4.0
 python-pptx==1.0.2
+qdrant-client==1.13.2
 redis==5.2.1
-referencing>=0.28.0,<0.31.0
+referencing==0.30.2
 regex==2024.11.6
 requests==2.32.3
 retry==0.9.2
 sentence-transformers==3.3.1
 tiktoken==0.8.0
 tokenizers==0.21.0
-torch==2.7.0
+torch==2.5.1
 tqdm==4.67.1
-transformers==4.51.3
+transformers==4.49.0
 typing-extensions==4.12.2
 typing-inspect==0.9.0
 tzdata==2024.2
@@ -79,7 +91,7 @@ urllib3==2.3.0
 vine==5.1.0
 wcwidth==0.2.13
 werkzeug==3.1.3
-yarl==1.20.0
-markdownify==1.1.0
+yarl==1.18.3
+markdownify==0.14.1
 tldextract==5.1.3
 websockets==14.1

application/retriever/brave_search.py

@@ -0,0 +1,112 @@
+import json
+
+from langchain_community.tools import BraveSearch
+
+from application.core.settings import settings
+from application.llm.llm_creator import LLMCreator
+from application.retriever.base import BaseRetriever
+
+
+class BraveRetSearch(BaseRetriever):
+
+    def __init__(
+        self,
+        source,
+        chat_history,
+        prompt,
+        chunks=2,
+        token_limit=150,
+        gpt_model="docsgpt",
+        user_api_key=None,
+        decoded_token=None,
+    ):
+        self.question = ""
+        self.source = source
+        self.chat_history = chat_history
+        self.prompt = prompt
+        self.chunks = chunks
+        self.gpt_model = gpt_model
+        self.token_limit = (
+            token_limit
+            if token_limit
+            < settings.MODEL_TOKEN_LIMITS.get(
+                self.gpt_model, settings.DEFAULT_MAX_HISTORY
+            )
+            else settings.MODEL_TOKEN_LIMITS.get(
+                self.gpt_model, settings.DEFAULT_MAX_HISTORY
+            )
+        )
+        self.user_api_key = user_api_key
+        self.decoded_token = decoded_token
+
+    def _get_data(self):
+        if self.chunks == 0:
+            docs = []
+        else:
+            search = BraveSearch.from_api_key(
+                api_key=settings.BRAVE_SEARCH_API_KEY,
+                search_kwargs={"count": int(self.chunks)},
+            )
+            results = search.run(self.question)
+            results = json.loads(results)
+
+            docs = []
+            for i in results:
+                try:
+                    title = i["title"]
+                    link = i["link"]
+                    snippet = i["snippet"]
+                    docs.append({"text": snippet, "title": title, "link": link})
+                except IndexError:
+                    pass
+        if settings.LLM_NAME == "llama.cpp":
+            docs = [docs[0]]
+
+        return docs
+
+    def gen(self):
+        docs = self._get_data()
+
+        # join all page_content together with a newline
+        docs_together = "\n".join([doc["text"] for doc in docs])
+        p_chat_combine = self.prompt.replace("{summaries}", docs_together)
+        messages_combine = [{"role": "system", "content": p_chat_combine}]
+        for doc in docs:
+            yield {"source": doc}
+
+        if len(self.chat_history) > 0:
+            for i in self.chat_history:
+                if "prompt" in i and "response" in i:
+                    messages_combine.append({"role": "user", "content": i["prompt"]})
+                    messages_combine.append(
+                        {"role": "assistant", "content": i["response"]}
+                    )
+        messages_combine.append({"role": "user", "content": self.question})
+
+        llm = LLMCreator.create_llm(
+            settings.LLM_NAME,
+            api_key=settings.API_KEY,
+            user_api_key=self.user_api_key,
+            decoded_token=self.decoded_token,
+        )
+
+        completion = llm.gen_stream(model=self.gpt_model, messages=messages_combine)
+        for line in completion:
+            yield {"answer": str(line)}
+
+    def search(self, query: str = ""):
+        if query:
+            self.question = query
+        return self._get_data()
+
+    def get_params(self):
+        return {
+            "question": self.question,
+            "source": self.source,
+            "chat_history": self.chat_history,
+            "prompt": self.prompt,
+            "chunks": self.chunks,
+            "token_limit": self.token_limit,
+            "gpt_model": self.gpt_model,
+            "user_api_key": self.user_api_key,
+        }

application/retriever/classic_rag.py

@@ -1,4 +1,3 @@
-import logging
 from application.core.settings import settings
 from application.llm.llm_creator import LLMCreator
 from application.retriever.base import BaseRetriever
@@ -16,7 +15,7 @@ class ClassicRAG(BaseRetriever):
         token_limit=150,
         gpt_model="docsgpt",
         user_api_key=None,
-        llm_name=settings.LLM_PROVIDER,
+        llm_name=settings.LLM_NAME,
         api_key=settings.API_KEY,
         decoded_token=None,
     ):
@@ -28,10 +27,10 @@ class ClassicRAG(BaseRetriever):
         self.token_limit = (
             token_limit
             if token_limit
-            < settings.LLM_TOKEN_LIMITS.get(
+            < settings.MODEL_TOKEN_LIMITS.get(
                 self.gpt_model, settings.DEFAULT_MAX_HISTORY
             )
-            else settings.LLM_TOKEN_LIMITS.get(
+            else settings.MODEL_TOKEN_LIMITS.get(
                 self.gpt_model, settings.DEFAULT_MAX_HISTORY
             )
         )
@@ -44,8 +43,8 @@ class ClassicRAG(BaseRetriever):
             user_api_key=self.user_api_key,
             decoded_token=decoded_token,
         )
-        self.vectorstore = source["active_docs"] if "active_docs" in source else None
         self.question = self._rephrase_query()
+        self.vectorstore = source["active_docs"] if "active_docs" in source else None
         self.decoded_token = decoded_token
 
     def _rephrase_query(self):
@@ -53,8 +52,6 @@ class ClassicRAG(BaseRetriever):
             not self.original_question
             or not self.chat_history
             or self.chat_history == []
-            or self.chunks == 0
-            or self.vectorstore is None
         ):
             return self.original_question
 
@@ -75,11 +72,11 @@ class ClassicRAG(BaseRetriever):
             print(f"Rephrased query: {rephrased_query}")
             return rephrased_query if rephrased_query else self.original_question
         except Exception as e:
-            logging.error(f"Error rephrasing query: {e}", exc_info=True)
+            print(f"Error rephrasing query: {e}")
             return self.original_question
 
     def _get_data(self):
-        if self.chunks == 0 or self.vectorstore is None:
+        if self.chunks == 0:
             docs = []
         else:
             docsearch = VectorCreator.create_vectorstore(

application/retriever/duckduck_search.py

@@ -0,0 +1,111 @@
+from langchain_community.tools import DuckDuckGoSearchResults
+from langchain_community.utilities import DuckDuckGoSearchAPIWrapper
+
+from application.core.settings import settings
+from application.llm.llm_creator import LLMCreator
+from application.retriever.base import BaseRetriever
+
+
+class DuckDuckSearch(BaseRetriever):
+
+    def __init__(
+        self,
+        source,
+        chat_history,
+        prompt,
+        chunks=2,
+        token_limit=150,
+        gpt_model="docsgpt",
+        user_api_key=None,
+        decoded_token=None,
+    ):
+        self.question = ""
+        self.source = source
+        self.chat_history = chat_history
+        self.prompt = prompt
+        self.chunks = chunks
+        self.gpt_model = gpt_model
+        self.token_limit = (
+            token_limit
+            if token_limit
+            < settings.MODEL_TOKEN_LIMITS.get(
+                self.gpt_model, settings.DEFAULT_MAX_HISTORY
+            )
+            else settings.MODEL_TOKEN_LIMITS.get(
+                self.gpt_model, settings.DEFAULT_MAX_HISTORY
+            )
+        )
+        self.user_api_key = user_api_key
+        self.decoded_token = decoded_token
+
+    def _get_data(self):
+        if self.chunks == 0:
+            docs = []
+        else:
+            wrapper = DuckDuckGoSearchAPIWrapper(max_results=self.chunks)
+            search = DuckDuckGoSearchResults(api_wrapper=wrapper, output_format="list")
+            results = search.run(self.question)
+
+            docs = []
+            for i in results:
+                try:
+                    docs.append(
+                        {
+                            "text": i.get("snippet", "").strip(),
+                            "title": i.get("title", "").strip(),
+                            "link": i.get("link", "").strip(),
+                        }
+                    )
+                except IndexError:
+                    pass
+        if settings.LLM_NAME == "llama.cpp":
+            docs = [docs[0]]
+
+        return docs
+
+    def gen(self):
+        docs = self._get_data()
+
+        # join all page_content together with a newline
+        docs_together = "\n".join([doc["text"] for doc in docs])
+        p_chat_combine = self.prompt.replace("{summaries}", docs_together)
+        messages_combine = [{"role": "system", "content": p_chat_combine}]
+        for doc in docs:
+            yield {"source": doc}
+
+        if len(self.chat_history) > 0:
+            for i in self.chat_history:
+                if "prompt" in i and "response" in i:
+                    messages_combine.append({"role": "user", "content": i["prompt"]})
+                    messages_combine.append(
+                        {"role": "assistant", "content": i["response"]}
+                    )
+        messages_combine.append({"role": "user", "content": self.question})
+
+        llm = LLMCreator.create_llm(
+            settings.LLM_NAME,
+            api_key=settings.API_KEY,
+            user_api_key=self.user_api_key,
+            decoded_token=self.decoded_token,
+        )
+
+        completion = llm.gen_stream(model=self.gpt_model, messages=messages_combine)
+        for line in completion:
+            yield {"answer": str(line)}
+
+    def search(self, query: str = ""):
+        if query:
+            self.question = query
+        return self._get_data()
+
+    def get_params(self):
+        return {
+            "question": self.question,
+            "source": self.source,
+            "chat_history": self.chat_history,
+            "prompt": self.prompt,
+            "chunks": self.chunks,
+            "token_limit": self.token_limit,
+            "gpt_model": self.gpt_model,
+            "user_api_key": self.user_api_key,
+        }

application/retriever/retriever_creator.py

@@ -1,16 +1,20 @@
 from application.retriever.classic_rag import ClassicRAG
+from application.retriever.duckduck_search import DuckDuckSearch
+from application.retriever.brave_search import BraveRetSearch
+
 
 
 class RetrieverCreator:
     retrievers = {
-        "classic": ClassicRAG,
-        "default": ClassicRAG,
+        'classic': ClassicRAG,
+        'duckduck_search': DuckDuckSearch,
+        'brave_search': BraveRetSearch,
+        'default': ClassicRAG
     }
 
     @classmethod
     def create_retriever(cls, type, *args, **kwargs):
-        retriever_type = (type or "default").lower()
-        retiever_class = cls.retrievers.get(retriever_type)
+        retiever_class = cls.retrievers.get(type.lower())
         if not retiever_class:
             raise ValueError(f"No retievers class found for type {type}")
-        return retiever_class(*args, **kwargs)
+        return retiever_class(*args, **kwargs)

application/storage/base.py

@@ -1,95 +0,0 @@
-"""Base storage class for file system abstraction."""
-
-from abc import ABC, abstractmethod
-from typing import BinaryIO, List, Callable
-
-
-class BaseStorage(ABC):
-    """Abstract base class for storage implementations."""
-
-    @abstractmethod
-    def save_file(self, file_data: BinaryIO, path: str, **kwargs) -> dict:
-        """
-        Save a file to storage.
-
-        Args:
-            file_data: File-like object containing the data
-            path: Path where the file should be stored
-
-        Returns:
-            dict: A dictionary containing metadata about the saved file, including:
-                - 'path': The path where the file was saved
-                - 'storage_type': The type of storage (e.g., 'local', 's3')
-                - Other storage-specific metadata (e.g., 'uri', 'bucket_name', etc.)
-        """
-        pass
-
-    @abstractmethod
-    def get_file(self, path: str) -> BinaryIO:
-        """
-        Retrieve a file from storage.
-
-        Args:
-            path: Path to the file
-
-        Returns:
-            BinaryIO: File-like object containing the file data
-        """
-        pass
-
-    @abstractmethod
-    def process_file(self, path: str, processor_func: Callable, **kwargs):
-        """
-        Process a file using the provided processor function.
-
-        This method handles the details of retrieving the file and providing
-        it to the processor function in an appropriate way based on the storage type.
-
-        Args:
-            path: Path to the file
-            processor_func: Function that processes the file
-            **kwargs: Additional arguments to pass to the processor function
-
-        Returns:
-            The result of the processor function
-        """
-        pass
-
-    @abstractmethod
-    def delete_file(self, path: str) -> bool:
-        """
-        Delete a file from storage.
-
-        Args:
-            path: Path to the file
-
-        Returns:
-            bool: True if deletion was successful
-        """
-        pass
-
-    @abstractmethod
-    def file_exists(self, path: str) -> bool:
-        """
-        Check if a file exists.
-
-        Args:
-            path: Path to the file
-
-        Returns:
-            bool: True if the file exists
-        """
-        pass
-
-    @abstractmethod
-    def list_files(self, directory: str) -> List[str]:
-        """
-        List all files in a directory.
-
-        Args:
-            directory: Directory path to list
-
-        Returns:
-            List[str]: List of file paths
-        """
-        pass

application/storage/local.py

@@ -1,103 +0,0 @@
-"""Local file system implementation."""
-import os
-import shutil
-from typing import BinaryIO, List, Callable
-
-from application.storage.base import BaseStorage
-
-
-class LocalStorage(BaseStorage):
-    """Local file system storage implementation."""
-
-    def __init__(self, base_dir: str = None):
-        """
-        Initialize local storage.
-
-        Args:
-            base_dir: Base directory for all operations. If None, uses current directory.
-        """
-        self.base_dir = base_dir or os.path.dirname(
-            os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
-        )
-
-    def _get_full_path(self, path: str) -> str:
-        """Get absolute path by combining base_dir and path."""
-        if os.path.isabs(path):
-            return path
-        return os.path.join(self.base_dir, path)
-
-    def save_file(self, file_data: BinaryIO, path: str) -> dict:
-        """Save a file to local storage."""
-        full_path = self._get_full_path(path)
-
-        os.makedirs(os.path.dirname(full_path), exist_ok=True)
-
-        if hasattr(file_data, 'save'):
-            file_data.save(full_path)
-        else:
-            with open(full_path, 'wb') as f:
-                shutil.copyfileobj(file_data, f)
-
-        return {
-            'storage_type': 'local'
-        }
-
-    def get_file(self, path: str) -> BinaryIO:
-        """Get a file from local storage."""
-        full_path = self._get_full_path(path)
-
-        if not os.path.exists(full_path):
-            raise FileNotFoundError(f"File not found: {full_path}")
-
-        return open(full_path, 'rb')
-
-    def delete_file(self, path: str) -> bool:
-        """Delete a file from local storage."""
-        full_path = self._get_full_path(path)
-
-        if not os.path.exists(full_path):
-            return False
-
-        os.remove(full_path)
-        return True
-
-    def file_exists(self, path: str) -> bool:
-        """Check if a file exists in local storage."""
-        full_path = self._get_full_path(path)
-        return os.path.exists(full_path)
-
-    def list_files(self, directory: str) -> List[str]:
-        """List all files in a directory in local storage."""
-        full_path = self._get_full_path(directory)
-
-        if not os.path.exists(full_path):
-            return []
-
-        result = []
-        for root, _, files in os.walk(full_path):
-            for file in files:
-                rel_path = os.path.relpath(os.path.join(root, file), self.base_dir)
-                result.append(rel_path)
-
-        return result
-
-    def process_file(self, path: str, processor_func: Callable, **kwargs):
-        """
-        Process a file using the provided processor function.
-
-        For local storage, we can directly pass the full path to the processor.
-
-        Args:
-            path: Path to the file
-            processor_func: Function that processes the file
-            **kwargs: Additional arguments to pass to the processor function
-
-        Returns:
-            The result of the processor function
-        """
-        full_path = self._get_full_path(path)
-
-        if not os.path.exists(full_path):
-            raise FileNotFoundError(f"File not found: {full_path}")
-
-        return processor_func(local_path=full_path, **kwargs)

application/storage/s3.py

@@ -1,132 +0,0 @@
-"""S3 storage implementation."""
-
-import io
-import os
-from typing import BinaryIO, Callable, List
-
-import boto3
-from application.core.settings import settings
-
-from application.storage.base import BaseStorage
-from botocore.exceptions import ClientError
-
-
-class S3Storage(BaseStorage):
-    """AWS S3 storage implementation."""
-
-    def __init__(self, bucket_name=None):
-        """
-        Initialize S3 storage.
-
-        Args:
-            bucket_name: S3 bucket name (optional, defaults to settings)
-        """
-        self.bucket_name = bucket_name or getattr(
-            settings, "S3_BUCKET_NAME", "docsgpt-test-bucket"
-        )
-
-        # Get credentials from settings
-
-        aws_access_key_id = getattr(settings, "SAGEMAKER_ACCESS_KEY", None)
-        aws_secret_access_key = getattr(settings, "SAGEMAKER_SECRET_KEY", None)
-        region_name = getattr(settings, "SAGEMAKER_REGION", None)
-
-        self.s3 = boto3.client(
-            "s3",
-            aws_access_key_id=aws_access_key_id,
-            aws_secret_access_key=aws_secret_access_key,
-            region_name=region_name,
-        )
-
-    def save_file(
-        self,
-        file_data: BinaryIO,
-        path: str,
-        storage_class: str = "INTELLIGENT_TIERING",
-        **kwargs,
-    ) -> dict:
-        """Save a file to S3 storage."""
-        self.s3.upload_fileobj(
-            file_data, self.bucket_name, path, ExtraArgs={"StorageClass": storage_class}
-        )
-
-        region = getattr(settings, "SAGEMAKER_REGION", None)
-
-        return {
-            "storage_type": "s3",
-            "bucket_name": self.bucket_name,
-            "uri": f"s3://{self.bucket_name}/{path}",
-            "region": region,
-        }
-
-    def get_file(self, path: str) -> BinaryIO:
-        """Get a file from S3 storage."""
-        if not self.file_exists(path):
-            raise FileNotFoundError(f"File not found: {path}")
-        file_obj = io.BytesIO()
-        self.s3.download_fileobj(self.bucket_name, path, file_obj)
-        file_obj.seek(0)
-        return file_obj
-
-    def delete_file(self, path: str) -> bool:
-        """Delete a file from S3 storage."""
-        try:
-            self.s3.delete_object(Bucket=self.bucket_name, Key=path)
-            return True
-        except ClientError:
-            return False
-
-    def file_exists(self, path: str) -> bool:
-        """Check if a file exists in S3 storage."""
-        try:
-            self.s3.head_object(Bucket=self.bucket_name, Key=path)
-            return True
-        except ClientError:
-            return False
-
-    def list_files(self, directory: str) -> List[str]:
-        """List all files in a directory in S3 storage."""
-        # Ensure directory ends with a slash if it's not empty
-
-        if directory and not directory.endswith("/"):
-            directory += "/"
-        result = []
-        paginator = self.s3.get_paginator("list_objects_v2")
-        pages = paginator.paginate(Bucket=self.bucket_name, Prefix=directory)
-
-        for page in pages:
-            if "Contents" in page:
-                for obj in page["Contents"]:
-                    result.append(obj["Key"])
-        return result
-
-    def process_file(self, path: str, processor_func: Callable, **kwargs):
-        """
-        Process a file using the provided processor function.
-
-        Args:
-            path: Path to the file
-            processor_func: Function that processes the file
-            **kwargs: Additional arguments to pass to the processor function
-
-        Returns:
-            The result of the processor function
-        """
-        import logging
-        import tempfile
-
-        if not self.file_exists(path):
-            raise FileNotFoundError(f"File not found in S3: {path}")
-        with tempfile.NamedTemporaryFile(
-            suffix=os.path.splitext(path)[1], delete=True
-        ) as temp_file:
-            try:
-                # Download the file from S3 to the temporary file
-
-                self.s3.download_fileobj(self.bucket_name, path, temp_file)
-                temp_file.flush()
-
-                return processor_func(local_path=temp_file.name, **kwargs)
-            except Exception as e:
-                logging.error(f"Error processing S3 file {path}: {e}", exc_info=True)
-                raise

application/storage/storage_creator.py

@@ -1,32 +0,0 @@
-"""Storage factory for creating different storage implementations."""
-from typing import Dict, Type
-
-from application.storage.base import BaseStorage
-from application.storage.local import LocalStorage
-from application.storage.s3 import S3Storage
-from application.core.settings import settings
-
-
-class StorageCreator:
-    storages: Dict[str, Type[BaseStorage]] = {
-        "local": LocalStorage,
-        "s3": S3Storage,
-    }
-    
-    _instance = None
-    
-    @classmethod
-    def get_storage(cls) -> BaseStorage:
-        if cls._instance is None:
-            storage_type = getattr(settings, "STORAGE_TYPE", "local")
-            cls._instance = cls.create_storage(storage_type)
-        
-        return cls._instance
-    
-    @classmethod
-    def create_storage(cls, type_name: str, *args, **kwargs) -> BaseStorage:
-        storage_class = cls.storages.get(type_name.lower())
-        if not storage_class:
-            raise ValueError(f"No storage implementation found for type {type_name}")
-        
-        return storage_class(*args, **kwargs)

application/usage.py

@@ -2,11 +2,10 @@ import sys
 from datetime import datetime
 
 from application.core.mongo_db import MongoDB
-from application.core.settings import settings
 from application.utils import num_tokens_from_object_or_list, num_tokens_from_string
 
 mongo = MongoDB.get_client()
-db = mongo[settings.MONGO_DB_NAME]
+db = mongo["docsgpt"]
 usage_collection = db["token_usage"]
 
 

application/utils.py

@@ -1,12 +1,8 @@
 import hashlib
-import os
 import re
-import uuid
 
 import tiktoken
 from flask import jsonify, make_response
-from werkzeug.utils import secure_filename
-from application.core.settings import settings
 
 
 _encoding = None
@@ -19,31 +15,6 @@ def get_encoding():
     return _encoding
 
 
-def safe_filename(filename):
-    """
-    Creates a safe filename that preserves the original extension.
-    Uses secure_filename, but ensures a proper filename is returned even with non-Latin characters.
-
-    Args:
-        filename (str): The original filename
-
-    Returns:
-        str: A safe filename that can be used for storage
-    """
-    if not filename:
-        return str(uuid.uuid4())
-
-    _, extension = os.path.splitext(filename)
-
-    safe_name = secure_filename(filename)
-
-    # If secure_filename returns just the extension or an empty string
-    if not safe_name or safe_name == extension.lstrip("."):
-        return f"{str(uuid.uuid4())}{extension}"
-
-    return safe_name
-
-
 def num_tokens_from_string(string: str) -> int:
     encoding = get_encoding()
     if isinstance(string, str):
@@ -80,7 +51,7 @@ def check_required_fields(data, required_fields):
             jsonify(
                 {
                     "success": False,
-                    "message": f"Missing required fields: {', '.join(missing_fields)}",
+                    "message": f"Missing fields: {', '.join(missing_fields)}",
                 }
             ),
             400,
@@ -88,29 +59,6 @@ def check_required_fields(data, required_fields):
     return None
 
 
-def validate_required_fields(data, required_fields):
-    missing_fields = []
-    empty_fields = []
-
-    for field in required_fields:
-        if field not in data:
-            missing_fields.append(field)
-        elif not data[field]:
-            empty_fields.append(field)
-
-    errors = []
-    if missing_fields:
-        errors.append(f"Missing required fields: {', '.join(missing_fields)}")
-    if empty_fields:
-        errors.append(f"Empty values in required fields: {', '.join(empty_fields)}")
-
-    if errors:
-        return make_response(
-            jsonify({"success": False, "message": " | ".join(errors)}), 400
-        )
-    return None
-
-
 def get_hash(data):
     return hashlib.md5(data.encode(), usedforsecurity=False).hexdigest()
 
@@ -126,8 +74,8 @@ def limit_chat_history(history, max_token_limit=None, gpt_model="docsgpt"):
         max_token_limit
         if max_token_limit
         and max_token_limit
-        < settings.LLM_TOKEN_LIMITS.get(gpt_model, settings.DEFAULT_MAX_HISTORY)
-        else settings.LLM_TOKEN_LIMITS.get(gpt_model, settings.DEFAULT_MAX_HISTORY)
+        < settings.MODEL_TOKEN_LIMITS.get(gpt_model, settings.DEFAULT_MAX_HISTORY)
+        else settings.MODEL_TOKEN_LIMITS.get(gpt_model, settings.DEFAULT_MAX_HISTORY)
     )
 
     if not history:
@@ -161,14 +109,3 @@ def validate_function_name(function_name):
     if not re.match(r"^[a-zA-Z0-9_-]+$", function_name):
         return False
     return True
-
-
-def generate_image_url(image_path):
-    strategy = getattr(settings, "URL_STRATEGY", "backend")
-    if strategy == "s3":
-        bucket_name = getattr(settings, "S3_BUCKET_NAME", "docsgpt-test-bucket")
-        region_name = getattr(settings, "SAGEMAKER_REGION", "eu-central-1")
-        return f"https://{bucket_name}.s3.{region_name}.amazonaws.com/{image_path}"
-    else:
-        base_url = getattr(settings, "API_URL", "http://localhost:7091")
-        return f"{base_url}/api/images/{image_path}"

application/vectorstore/elasticsearch.py

@@ -1,6 +1,9 @@
 from application.vectorstore.base import BaseVectorStore
 from application.core.settings import settings
 from application.vectorstore.document_class import Document
+import elasticsearch
+
+
 
 
 class ElasticsearchStore(BaseVectorStore):
@@ -23,7 +26,8 @@ class ElasticsearchStore(BaseVectorStore):
             else:
                 raise ValueError("Please provide either elasticsearch_url or cloud_id.")
 
-            import elasticsearch
+            
+
             ElasticsearchStore._es_connection = elasticsearch.Elasticsearch(**connection_params)
             
         self.docsearch = ElasticsearchStore._es_connection
@@ -151,6 +155,8 @@ class ElasticsearchStore(BaseVectorStore):
         **kwargs,
         ):
         
+        from elasticsearch.helpers import BulkIndexError, bulk
+
         bulk_kwargs = bulk_kwargs or {}
         import uuid
         embeddings = []
@@ -183,7 +189,6 @@ class ElasticsearchStore(BaseVectorStore):
 
 
         if len(requests) > 0:
-            from elasticsearch.helpers import BulkIndexError, bulk
             try:
                 success, failed = bulk(
                     self._es_connection,

application/vectorstore/faiss.py

@@ -1,19 +1,17 @@
 import os
-import tempfile
 
 from langchain_community.vectorstores import FAISS
 
 from application.core.settings import settings
 from application.parser.schema.base import Document
 from application.vectorstore.base import BaseVectorStore
-from application.storage.storage_creator import StorageCreator
 
 
 def get_vectorstore(path: str) -> str:
     if path:
-        vectorstore = f"indexes/{path}"
+        vectorstore = os.path.join("application", "indexes", path)
     else:
-        vectorstore = "indexes"
+        vectorstore = os.path.join("application")
     return vectorstore
 
 
@@ -23,40 +21,16 @@ class FaissStore(BaseVectorStore):
         self.source_id = source_id
         self.path = get_vectorstore(source_id)
         self.embeddings = self._get_embeddings(settings.EMBEDDINGS_NAME, embeddings_key)
-        self.storage = StorageCreator.get_storage()
 
         try:
             if docs_init:
                 self.docsearch = FAISS.from_documents(docs_init, self.embeddings)
             else:
-                with tempfile.TemporaryDirectory() as temp_dir:
-                    faiss_path = f"{self.path}/index.faiss"
-                    pkl_path = f"{self.path}/index.pkl"
-
-                    if not self.storage.file_exists(
-                        faiss_path
-                    ) or not self.storage.file_exists(pkl_path):
-                        raise FileNotFoundError(
-                            f"Index files not found in storage at {self.path}"
-                        )
-
-                    faiss_file = self.storage.get_file(faiss_path)
-                    pkl_file = self.storage.get_file(pkl_path)
-
-                    local_faiss_path = os.path.join(temp_dir, "index.faiss")
-                    local_pkl_path = os.path.join(temp_dir, "index.pkl")
-
-                    with open(local_faiss_path, "wb") as f:
-                        f.write(faiss_file.read())
-
-                    with open(local_pkl_path, "wb") as f:
-                        f.write(pkl_file.read())
-
-                    self.docsearch = FAISS.load_local(
-                        temp_dir, self.embeddings, allow_dangerous_deserialization=True
-                    )
-        except Exception as e:
-            raise Exception(f"Error loading FAISS index: {str(e)}")
+                self.docsearch = FAISS.load_local(
+                    self.path, self.embeddings, allow_dangerous_deserialization=True
+                )
+        except Exception:
+            raise
 
         self.assert_embedding_dimensions(self.embeddings)
 

application/vectorstore/mongodb.py

@@ -1,4 +1,3 @@
-import logging
 from application.core.settings import settings
 from application.vectorstore.base import BaseVectorStore
 from application.vectorstore.document_class import Document
@@ -147,7 +146,7 @@ class MongoDBVectorStore(BaseVectorStore):
 
             return chunks
         except Exception as e:
-            logging.error(f"Error getting chunks: {e}", exc_info=True)
+            print(f"Error getting chunks: {e}")
             return []
 
     def add_chunk(self, text, metadata=None):
@@ -173,5 +172,5 @@ class MongoDBVectorStore(BaseVectorStore):
             result = self._collection.delete_one({"_id": object_id})
             return result.deleted_count > 0
         except Exception as e:
-            logging.error(f"Error deleting chunk: {e}", exc_info=True)
+            print(f"Error deleting chunk: {e}")
             return False

application/vectorstore/qdrant.py

@@ -1,12 +1,11 @@
+from langchain_community.vectorstores.qdrant import Qdrant
 from application.vectorstore.base import BaseVectorStore
 from application.core.settings import settings
+from qdrant_client import models
 
 
 class QdrantStore(BaseVectorStore):
     def __init__(self, source_id: str = "", embeddings_key: str = "embeddings"):
-        from qdrant_client import models
-        from langchain_community.vectorstores.qdrant import Qdrant
-
         self._filter = models.Filter(
             must=[
                 models.FieldCondition(

application/worker.py

@@ -1,37 +1,25 @@
-import datetime
-import json
 import logging
-import mimetypes
 import os
 import shutil
 import string
-import tempfile
 import zipfile
-
 from collections import Counter
 from urllib.parse import urljoin
 
 import requests
-from bson.dbref import DBRef
 from bson.objectid import ObjectId
 
-from application.agents.agent_creator import AgentCreator
-from application.api.answer.routes import get_prompt
-
 from application.core.mongo_db import MongoDB
 from application.core.settings import settings
-from application.parser.chunking import Chunker
-from application.parser.embedding_pipeline import embed_and_store_documents
 from application.parser.file.bulk import SimpleDirectoryReader
+from application.parser.embedding_pipeline import embed_and_store_documents
 from application.parser.remote.remote_creator import RemoteCreator
 from application.parser.schema.base import Document
-from application.retriever.retriever_creator import RetrieverCreator
-
-from application.storage.storage_creator import StorageCreator
-from application.utils import count_tokens_docs, num_tokens_from_string
+from application.parser.chunking import Chunker
+from application.utils import count_tokens_docs
 
 mongo = MongoDB.get_client()
-db = mongo[settings.MONGO_DB_NAME]
+db = mongo["docsgpt"]
 sources_collection = db["sources"]
 
 # Constants
@@ -39,22 +27,18 @@ MIN_TOKENS = 150
 MAX_TOKENS = 1250
 RECURSION_DEPTH = 2
 
-
 # Define a function to extract metadata from a given filename.
 def metadata_from_filename(title):
     return {"title": title}
 
-
 # Define a function to generate a random string of a given length.
 def generate_random_string(length):
     return "".join([string.ascii_letters[i % 52] for i in range(length)])
 
-
 current_dir = os.path.dirname(
     os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
 )
 
-
 def extract_zip_recursive(zip_path, extract_to, current_depth=0, max_depth=5):
     """
     Recursively extract zip files with a limit on recursion depth.
@@ -74,7 +58,7 @@ def extract_zip_recursive(zip_path, extract_to, current_depth=0, max_depth=5):
             zip_ref.extractall(extract_to)
         os.remove(zip_path)  # Remove the zip file after extracting
     except Exception as e:
-        logging.error(f"Error extracting zip file {zip_path}: {e}", exc_info=True)
+        logging.error(f"Error extracting zip file {zip_path}: {e}")
         return
 
     # Check for nested zip files and extract them
@@ -85,7 +69,6 @@ def extract_zip_recursive(zip_path, extract_to, current_depth=0, max_depth=5):
                 file_path = os.path.join(root, file)
                 extract_zip_recursive(file_path, root, current_depth + 1, max_depth)
 
-
 def download_file(url, params, dest_path):
     try:
         response = requests.get(url, params=params)
@@ -96,7 +79,6 @@ def download_file(url, params, dest_path):
         logging.error(f"Error downloading file: {e}")
         raise
 
-
 def upload_index(full_path, file_data):
     try:
         if settings.VECTOR_STORE == "faiss":
@@ -105,9 +87,7 @@ def upload_index(full_path, file_data):
                 "file_pkl": open(full_path + "/index.pkl", "rb"),
             }
             response = requests.post(
-                urljoin(settings.API_URL, "/api/upload_index"),
-                files=files,
-                data=file_data,
+                urljoin(settings.API_URL, "/api/upload_index"), files=files, data=file_data
             )
         else:
             response = requests.post(
@@ -122,79 +102,9 @@ def upload_index(full_path, file_data):
             for file in files.values():
                 file.close()
 
-
-def run_agent_logic(agent_config, input_data):
-    try:
-        source = agent_config.get("source")
-        retriever = agent_config.get("retriever", "classic")
-        if isinstance(source, DBRef):
-            source_doc = db.dereference(source)
-            source = str(source_doc["_id"])
-            retriever = source_doc.get("retriever", agent_config.get("retriever"))
-        else:
-            source = {}
-        source = {"active_docs": source}
-        chunks = int(agent_config.get("chunks", 2))
-        prompt_id = agent_config.get("prompt_id", "default")
-        user_api_key = agent_config["key"]
-        agent_type = agent_config.get("agent_type", "classic")
-        decoded_token = {"sub": agent_config.get("user")}
-        prompt = get_prompt(prompt_id)
-        agent = AgentCreator.create_agent(
-            agent_type,
-            endpoint="webhook",
-            llm_name=settings.LLM_PROVIDER,
-            gpt_model=settings.LLM_NAME,
-            api_key=settings.API_KEY,
-            user_api_key=user_api_key,
-            prompt=prompt,
-            chat_history=[],
-            decoded_token=decoded_token,
-            attachments=[],
-        )
-        retriever = RetrieverCreator.create_retriever(
-            retriever,
-            source=source,
-            chat_history=[],
-            prompt=prompt,
-            chunks=chunks,
-            token_limit=settings.DEFAULT_MAX_HISTORY,
-            gpt_model=settings.LLM_NAME,
-            user_api_key=user_api_key,
-            decoded_token=decoded_token,
-        )
-        answer = agent.gen(query=input_data, retriever=retriever)
-        response_full = ""
-        thought = ""
-        source_log_docs = []
-        tool_calls = []
-
-        for line in answer:
-            if "answer" in line:
-                response_full += str(line["answer"])
-            elif "sources" in line:
-                source_log_docs.extend(line["sources"])
-            elif "tool_calls" in line:
-                tool_calls.extend(line["tool_calls"])
-            elif "thought" in line:
-                thought += line["thought"]
-
-        result = {
-            "answer": response_full,
-            "sources": source_log_docs,
-            "tool_calls": tool_calls,
-            "thought": thought,
-        }
-        logging.info(f"Agent response: {result}")
-        return result
-    except Exception as e:
-        logging.error(f"Error in run_agent_logic: {e}", exc_info=True)
-        raise
-
-
 # Define the main function for ingesting and processing documents.
 def ingest_worker(
-    self, directory, formats, job_name, filename, user, dir_name=None, user_dir=None, retriever="classic"
+    self, directory, formats, name_job, filename, user, retriever="classic"
 ):
     """
     Ingest and process documents.
@@ -203,11 +113,9 @@ def ingest_worker(
         self: Reference to the instance of the task.
         directory (str): Specifies the directory for ingesting ('inputs' or 'temp').
         formats (list of str): List of file extensions to consider for ingestion (e.g., [".rst", ".md"]).
-        job_name (str): Name of the job for this ingestion task (original, unsanitized).
+        name_job (str): Name of the job for this ingestion task.
         filename (str): Name of the file to be ingested.
-        user (str): Identifier for the user initiating the ingestion (original, unsanitized).
-        dir_name (str, optional): Sanitized directory name for filesystem operations.
-        user_dir (str, optional): Sanitized user ID for filesystem operations.
+        user (str): Identifier for the user initiating the ingestion.
         retriever (str): Type of retriever to use for processing the documents.
 
     Returns:
@@ -218,99 +126,72 @@ def ingest_worker(
     limit = None
     exclude = True
     sample = False
-    
-    storage = StorageCreator.get_storage()
+    full_path = os.path.join(directory, user, name_job)
 
-    full_path = os.path.join(directory, user_dir, dir_name)
-    source_file_path = os.path.join(full_path, filename)
+    logging.info(f"Ingest file: {full_path}", extra={"user": user, "job": name_job})
+    file_data = {"name": name_job, "file": filename, "user": user}
 
-    logging.info(f"Ingest file: {full_path}", extra={"user": user, "job": job_name})
+    if not os.path.exists(full_path):
+        os.makedirs(full_path)
+    download_file(urljoin(settings.API_URL, "/api/download"), file_data, os.path.join(full_path, filename))
 
-    # Create temporary working directory
-    with tempfile.TemporaryDirectory() as temp_dir:
-        try:
-            os.makedirs(temp_dir, exist_ok=True)
+    # check if file is .zip and extract it
+    if filename.endswith(".zip"):
+        extract_zip_recursive(
+            os.path.join(full_path, filename), full_path, 0, RECURSION_DEPTH
+        )
 
-            # Download file from storage to temp directory
-            temp_file_path = os.path.join(temp_dir, filename)
-            file_data = storage.get_file(source_file_path)
+    self.update_state(state="PROGRESS", meta={"current": 1})
 
-            with open(temp_file_path, "wb") as f:
-                f.write(file_data.read())
+    raw_docs = SimpleDirectoryReader(
+        input_dir=full_path,
+        input_files=input_files,
+        recursive=recursive,
+        required_exts=formats,
+        num_files_limit=limit,
+        exclude_hidden=exclude,
+        file_metadata=metadata_from_filename,
+    ).load_data()
 
-            self.update_state(state="PROGRESS", meta={"current": 1})
+    chunker = Chunker(
+        chunking_strategy="classic_chunk",
+        max_tokens=MAX_TOKENS,
+        min_tokens=MIN_TOKENS,
+        duplicate_headers=False
+    )
+    raw_docs = chunker.chunk(documents=raw_docs)
 
-            # Handle zip files
-            if filename.endswith(".zip"):
-                logging.info(f"Extracting zip file: {filename}")
-                extract_zip_recursive(
-                    temp_file_path, temp_dir, current_depth=0, max_depth=RECURSION_DEPTH
-                )
+    docs = [Document.to_langchain_format(raw_doc) for raw_doc in raw_docs]
+    id = ObjectId()
 
-            if sample:
-                logging.info(f"Sample mode enabled. Using {limit} documents.")
+    embed_and_store_documents(docs, full_path, id, self)
+    tokens = count_tokens_docs(docs)
+    self.update_state(state="PROGRESS", meta={"current": 100})
 
-            reader = SimpleDirectoryReader(
-                input_dir=temp_dir,
-                input_files=input_files,
-                recursive=recursive,
-                required_exts=formats,
-                exclude_hidden=exclude,
-                file_metadata=metadata_from_filename,
-            )
-            raw_docs = reader.load_data()
+    if sample:
+        for i in range(min(5, len(raw_docs))):
+            logging.info(f"Sample document {i}: {raw_docs[i]}")
 
-            chunker = Chunker(
-                chunking_strategy="classic_chunk",
-                max_tokens=MAX_TOKENS,
-                min_tokens=MIN_TOKENS,
-                duplicate_headers=False,
-            )
-            raw_docs = chunker.chunk(documents=raw_docs)
+    file_data.update({
+        "tokens": tokens,
+        "retriever": retriever,
+        "id": str(id),
+        "type": "local",
+    })
+    upload_index(full_path, file_data)
 
-            docs = [Document.to_langchain_format(raw_doc) for raw_doc in raw_docs]
-
-            id = ObjectId()
-
-            vector_store_path = os.path.join(temp_dir, "vector_store")
-            os.makedirs(vector_store_path, exist_ok=True)
-
-            embed_and_store_documents(docs, vector_store_path, id, self)
-
-            tokens = count_tokens_docs(docs)
-
-            self.update_state(state="PROGRESS", meta={"current": 100})
-
-            if sample:
-                for i in range(min(5, len(raw_docs))):
-                    logging.info(f"Sample document {i}: {raw_docs[i]}")
-            file_data = {
-                "name": job_name,  # Use original job_name
-                "file": filename,
-                "user": user,  # Use original user
-                "tokens": tokens,
-                "retriever": retriever,
-                "id": str(id),
-                "type": "local",
-                "original_file_path": source_file_path,
-            }
-
-            upload_index(vector_store_path, file_data)
-
-        except Exception as e:
-            logging.error(f"Error in ingest_worker: {e}", exc_info=True)
-            raise
+    # delete local
+    shutil.rmtree(full_path)
 
     return {
         "directory": directory,
         "formats": formats,
-        "name_job": job_name,  # Use original job_name
+        "name_job": name_job,
         "filename": filename,
-        "user": user,  # Use original user
+        "user": user,
         "limited": False,
     }
 
-
 def remote_worker(
     self,
     source_data,
@@ -322,7 +203,7 @@ def remote_worker(
     sync_frequency="never",
     operation_mode="upload",
     doc_id=None,
-):
+):  
     full_path = os.path.join(directory, user, name_job)
     if not os.path.exists(full_path):
         os.makedirs(full_path)
@@ -337,7 +218,7 @@ def remote_worker(
             chunking_strategy="classic_chunk",
             max_tokens=MAX_TOKENS,
             min_tokens=MIN_TOKENS,
-            duplicate_headers=False,
+            duplicate_headers=False
         )
         docs = chunker.chunk(documents=raw_docs)
         docs = [Document.to_langchain_format(raw_doc) for raw_doc in raw_docs]
@@ -379,7 +260,6 @@ def remote_worker(
     logging.info("remote_worker task completed successfully")
     return {"urls": source_data, "name_job": name_job, "user": user, "limited": False}
 
-
 def sync(
     self,
     source_data,
@@ -405,11 +285,10 @@ def sync(
             doc_id,
         )
     except Exception as e:
-        logging.error(f"Error during sync: {e}", exc_info=True)
+        logging.error(f"Error during sync: {e}")
         return {"status": "error", "error": str(e)}
     return {"status": "success"}
 
-
 def sync_worker(self, frequency):
     sync_counts = Counter()
     sources = sources_collection.find()
@@ -433,122 +312,3 @@ def sync_worker(self, frequency):
         key: sync_counts[key]
         for key in ["total_sync_count", "sync_success", "sync_failure"]
     }
-
-
-def attachment_worker(self, file_info, user):
-    """
-    Process and store a single attachment without vectorization.
-    """
-
-    mongo = MongoDB.get_client()
-    db = mongo[settings.MONGO_DB_NAME]
-    attachments_collection = db["attachments"]
-
-    filename = file_info["filename"]
-    attachment_id = file_info["attachment_id"]
-    relative_path = file_info["path"]
-    metadata = file_info.get("metadata", {})
-
-    try:
-        self.update_state(state="PROGRESS", meta={"current": 10})
-        storage = StorageCreator.get_storage()
-
-        self.update_state(
-            state="PROGRESS", meta={"current": 30, "status": "Processing content"}
-        )
-
-        content = storage.process_file(
-            relative_path,
-            lambda local_path, **kwargs: SimpleDirectoryReader(
-                input_files=[local_path], exclude_hidden=True, errors="ignore"
-            )
-            .load_data()[0]
-            .text,
-        )
-
-        token_count = num_tokens_from_string(content)
-
-        self.update_state(
-            state="PROGRESS", meta={"current": 80, "status": "Storing in database"}
-        )
-
-        mime_type = mimetypes.guess_type(filename)[0] or "application/octet-stream"
-
-        doc_id = ObjectId(attachment_id)
-        attachments_collection.insert_one(
-            {
-                "_id": doc_id,
-                "user": user,
-                "path": relative_path,
-                "filename": filename,
-                "content": content,
-                "token_count": token_count,
-                "mime_type": mime_type,
-                "date": datetime.datetime.now(),
-                "metadata": metadata,
-            }
-        )
-
-        logging.info(
-            f"Stored attachment with ID: {attachment_id}", extra={"user": user}
-        )
-
-        self.update_state(state="PROGRESS", meta={"current": 100, "status": "Complete"})
-
-        return {
-            "filename": filename,
-            "path": relative_path,
-            "token_count": token_count,
-            "attachment_id": attachment_id,
-            "mime_type": mime_type,
-            "metadata": metadata,
-        }
-
-    except Exception as e:
-        logging.error(
-            f"Error processing file {filename}: {e}",
-            extra={"user": user},
-            exc_info=True,
-        )
-        raise
-
-
-def agent_webhook_worker(self, agent_id, payload):
-    """
-    Process the webhook payload for an agent.
-
-    Args:
-        self: Reference to the instance of the task.
-        agent_id (str): Unique identifier for the agent.
-        payload (dict): The payload data from the webhook.
-
-    Returns:
-        dict: Information about the processed webhook.
-    """
-    mongo = MongoDB.get_client()
-    db = mongo["docsgpt"]
-    agents_collection = db["agents"]
-
-    self.update_state(state="PROGRESS", meta={"current": 1})
-    try:
-        agent_oid = ObjectId(agent_id)
-        agent_config = agents_collection.find_one({"_id": agent_oid})
-        if not agent_config:
-            raise ValueError(f"Agent with ID {agent_id} not found.")
-        input_data = json.dumps(payload)
-    except Exception as e:
-        logging.error(f"Error processing agent webhook: {e}", exc_info=True)
-        return {"status": "error", "error": str(e)}
-
-    self.update_state(state="PROGRESS", meta={"current": 50})
-    try:
-        result = run_agent_logic(agent_config, input_data)
-    except Exception as e:
-        logging.error(f"Error running agent logic: {e}", exc_info=True)
-        return {"status": "error", "error": str(e)}
-    finally:
-        self.update_state(state="PROGRESS", meta={"current": 100})
-        logging.info(
-            f"Webhook processed for agent {agent_id}", extra={"agent_id": agent_id}
-        )
-        return {"status": "success", "result": result}

deployment/docker-compose-dev.yaml

@@ -1,4 +1,3 @@
-name: docsgpt-oss
 services:
 
   redis:

deployment/docker-compose.yaml

@@ -1,4 +1,3 @@
-name: docsgpt-oss
 services:
   frontend:
     build: ../frontend
@@ -13,46 +12,39 @@ services:
       - backend
 
   backend:
-    user: root
     build: ../application
     environment:
       - API_KEY=$API_KEY
       - EMBEDDINGS_KEY=$API_KEY
-      - LLM_PROVIDER=$LLM_PROVIDER
       - LLM_NAME=$LLM_NAME
       - CELERY_BROKER_URL=redis://redis:6379/0
       - CELERY_RESULT_BACKEND=redis://redis:6379/1
       - MONGO_URI=mongodb://mongo:27017/docsgpt
       - CACHE_REDIS_URL=redis://redis:6379/2
       - OPENAI_BASE_URL=$OPENAI_BASE_URL
+      - MODEL_NAME=$MODEL_NAME
     ports:
       - "7091:7091"
     volumes:
-      - ../application/indexes:/app/indexes
-      - ../application/inputs:/app/inputs
-      - ../application/vectors:/app/vectors
+      - ../application/indexes:/app/application/indexes
+      - ../application/inputs:/app/application/inputs
+      - ../application/vectors:/app/application/vectors
     depends_on:
       - redis
       - mongo
 
   worker:
-    user: root
     build: ../application
-    command: celery -A application.app.celery worker -l INFO -B
+    command: celery -A application.app.celery worker -l INFO --pool=gevent -B
     environment:
       - API_KEY=$API_KEY
       - EMBEDDINGS_KEY=$API_KEY
-      - LLM_PROVIDER=$LLM_PROVIDER
       - LLM_NAME=$LLM_NAME
       - CELERY_BROKER_URL=redis://redis:6379/0
       - CELERY_RESULT_BACKEND=redis://redis:6379/1
       - MONGO_URI=mongodb://mongo:27017/docsgpt
       - API_URL=http://backend:7091
       - CACHE_REDIS_URL=redis://redis:6379/2
-    volumes:
-      - ../application/indexes:/app/indexes
-      - ../application/inputs:/app/inputs
-      - ../application/vectors:/app/vectors
     depends_on:
       - redis
       - mongo

deployment/k8s/docsgpt-secrets.yaml

@@ -4,7 +4,7 @@ metadata:
   name: docsgpt-secrets
 type: Opaque
 data:
-  LLM_PROVIDER: ZG9jc2dwdA==
+  LLM_NAME: ZG9jc2dwdA==
   INTERNAL_KEY: aW50ZXJuYWw=
   CELERY_BROKER_URL: cmVkaXM6Ly9yZWRpcy1zZXJ2aWNlOjYzNzkvMA==
   CELERY_RESULT_BACKEND: cmVkaXM6Ly9yZWRpcy1zZXJ2aWNlOjYzNzkvMA==

docs/components/ToolCards.jsx

@@ -1,117 +0,0 @@
-import Image from 'next/image';
-
-const iconMap = {
-    'API Tool': '/toolIcons/tool_api_tool.svg',
-    'Brave Search Tool': '/toolIcons/tool_brave.svg',
-    'Cryptoprice Tool': '/toolIcons/tool_cryptoprice.svg',
-    'Ntfy Tool': '/toolIcons/tool_ntfy.svg',
-    'PostgreSQL Tool': '/toolIcons/tool_postgres.svg',
-    'Read Webpage Tool': '/toolIcons/tool_read_webpage.svg',
-    'Telegram Tool': '/toolIcons/tool_telegram.svg'
-};
-
-
-export function ToolCards({ items }) {
-    return (
-        <>
-            <div className="tool-cards">
-                {items.map(({ title, link, description }) => {
-                    const isExternal = link.startsWith('https://');
-                    const iconSrc = iconMap[title] || '/default-icon.png'; // Default icon if not found
-
-                    return (
-                        <div
-                            key={title}
-                            className={`card${isExternal ? ' external' : ''}`}
-                        >
-                            <a href={link} target={isExternal ? '_blank' : undefined} rel="noopener noreferrer" className="card-link-wrapper">
-                                <div className="card-icon-container">
-                                    {iconSrc && <div className="card-icon"><Image src={iconSrc} alt={title} width={32} height={32} /></div>} {/* Reduced icon size */}
-                                </div>
-                                <h3 className="card-title">{title}</h3>
-                                {description && <p className="card-description">{description}</p>}
-                                {/* Card URL element removed from here */}
-                            </a>
-                        </div>
-                    );
-                })}
-            </div>
-
-            <style jsx>{`
-               .tool-cards {
-                    margin-top: 24px;
-                    display: grid;
-                    grid-template-columns: 1fr;
-                    gap: 16px;
-                }
-                @media (min-width: 768px) {
-                    .tool-cards {
-                        grid-template-columns: 1fr 1fr; /* Keeps two columns on wider screens */
-                    }
-                }
-                .card {
-                    background-color: #222222;
-                    border-radius: 8px;
-                    padding: 16px; /* Existing padding */
-                    transition: background-color 0.3s;
-                    position: relative;
-                    color: #ffffff;
-                    display: flex; /* Using flex to help with alignment */
-                    flex-direction: column;
-                    /* align-items: center; // Alignment for items inside card-link-wrapper is better */
-                    /* justify-content: center; // We want content to flow from top */
-                    height: 100%; /* Fill the height of the grid cell, ensures cards in a row are same height */
-                }
-                .card:hover {
-                    background-color: #333333;
-                }
-                .card.external::after {
-                    content: "↗";
-                    position: absolute;
-                    top: 12px;
-                    right: 12px;
-                    color: #ffffff;
-                    font-size: 0.7em;
-                    opacity: 0.8;
-                }
-                .card-link-wrapper {
-                    display: flex;
-                    flex-direction: column;
-                    align-items:center; /* Centers icon, title, description horizontally */
-                    text-align: center; /* Ensures text within p and h3 is centered */
-                    color: inherit;
-                    text-decoration: none;
-                    width:100%;
-                    height: 100%; /* Make the link wrapper take full card height */
-                    justify-content: flex-start; /* Align content to the top */
-                }
-               .card-icon-container{
-                    display:flex;
-                    justify-content:center;
-                    width: 100%;
-                    margin-top: 8px; /* Added some margin at the top if needed */
-                    margin-bottom: 12px; /* Increased space between icon and title */
-               }
-                .card-icon {
-                   display: block;
-                   /* margin: 0 auto; // Center handled by card-icon-container */
-                }
-                .card-title {
-                    font-weight: 600;
-                    margin-bottom: 8px; /* Increased space below title */
-                    font-size: 16px; /* Consider increasing slightly if descriptions are longer e.g. 17px or 18px */
-                    color: #f0f0f0;
-                }
-                .card-description {
-                    /* margin-bottom: 0; // Original value */
-                    font-size: 14px; /* Slightly increased font size for better readability */
-                    color: #aaaaaa;
-                    line-height: 1.5; /* Slightly increased line height */
-                    flex-grow: 1; /* Allows description to take available space */
-                    overflow-y: auto; /* Adds scroll if description is too long, though ideally content fits */
-                    padding-bottom: 8px; /* Add some padding at the bottom of the description area */
-                }
-            `}</style>
-        </>
-    );
-}

docs/package.json

@@ -7,8 +7,8 @@
   "license": "MIT",
   "dependencies": {
     "@vercel/analytics": "^1.1.1",
-    "docsgpt-react": "^0.5.1",
-    "next": "^15.3.3",
+    "docsgpt-react": "^0.5.0",
+    "next": "^14.2.26",
     "nextra": "^2.13.2",
     "nextra-theme-docs": "^2.13.2",
     "react": "^18.2.0",

docs/pages/Agents/_meta.json

@@ -1,14 +0,0 @@
-{
-  "basics": {
-    "title": "🤖 Agent Basics",
-    "href": "/Agents/basics"
-  },
-  "api": {
-    "title": "🔌 Agent API",
-    "href": "/Agents/api"
-  },
-  "webhooks": {
-    "title": "🪝 Agent Webhooks",
-    "href": "/Agents/webhooks"
-  }
-}

docs/pages/Agents/api.mdx

@@ -1,227 +0,0 @@
----
-title: Interacting with Agents via API
-description: Learn how to programmatically interact with DocsGPT Agents using the streaming and non-streaming API endpoints.
----
-
-import { Callout, Tabs } from 'nextra/components';
-
-# Interacting with Agents via API
-
-DocsGPT Agents can be accessed programmatically through a dedicated API, allowing you to integrate their specialized capabilities into your own applications, scripts, and workflows. This guide covers the two primary methods for interacting with an agent: the streaming API for real-time responses and the non-streaming API for a single, consolidated answer.
-
-When you use an API key generated for a specific agent, you do not need to pass `prompt`, `tools` etc. The agent's configuration (including its prompt, selected tools, and knowledge sources) is already associated with its unique API key. 
-
-### API Endpoints
-
-- **Non-Streaming:** `http://localhost:7091/api/answer`
-- **Streaming:** `http://localhost:7091/stream`
-
-<Callout type="info">
-For DocsGPT Cloud, use `https://gptcloud.arc53.com/` as the base URL.
-</Callout>
-
-For more technical details, you can explore the API swagger documentation available for the cloud version or your local instance.
-
----
-
-## Non-Streaming API (`/api/answer`)
-
-This is a standard synchronous endpoint. It waits for the agent to fully process the request and returns a single JSON object with the complete answer. This is the simplest method and is ideal for backend processes where a real-time feed is not required.
-
-### Request
-
--   **Endpoint:** `/api/answer`
--   **Method:** `POST`
--   **Payload:**
-    -   `question` (string, required): The user's query or input for the agent.
-    -   `api_key` (string, required): The unique API key for the agent you wish to interact with.
-    -   `history` (string, optional): A JSON string representing the conversation history, e.g., `[{\"prompt\": \"first question\", \"answer\": \"first answer\"}]`.
-
-### Response
-
-A single JSON object containing:
--   `answer`: The complete, final answer from the agent.
--   `sources`: A list of sources the agent consulted.
--   `conversation_id`: The unique ID for the interaction.
-
-### Examples
-
-<Tabs items={['cURL', 'Python', 'JavaScript']}>
-  <Tabs.Tab>
-    ```bash
-    curl -X POST http://localhost:7091/api/answer \
-    -H "Content-Type: application/json" \
-    -d '{
-        "question": "your question here",
-        "api_key": "your_agent_api_key"
-    }'
-    ```
-  </Tabs.Tab>
-  <Tabs.Tab>
-    ```python
-    import requests
-
-    API_URL = "http://localhost:7091/api/answer"
-    API_KEY = "your_agent_api_key"
-    QUESTION = "your question here"
-
-    response = requests.post(
-        API_URL,
-        json={"question": QUESTION, "api_key": API_KEY}
-    )
-
-    if response.status_code == 200:
-        print(response.json())
-    else:
-        print(f"Error: {response.status_code}")
-        print(response.text)
-    ```
-  </Tabs.Tab>
-  <Tabs.Tab>
-    ```javascript
-    const apiUrl = 'http://localhost:7091/api/answer';
-    const apiKey = 'your_agent_api_key';
-    const question = 'your question here';
-
-    async function getAnswer() {
-    try {
-        const response = await fetch(apiUrl, {
-        method: 'POST',
-        headers: {
-            'Content-Type': 'application/json',
-        },
-        body: JSON.stringify({ question, api_key: apiKey }),
-        });
-
-        if (!response.ok) {
-        throw new Error(`HTTP error! Status: ${response.status}`);
-        }
-
-        const data = await response.json();
-        console.log(data);
-    } catch (error) {
-        console.error("Failed to fetch answer:", error);
-    }
-    }
-
-    getAnswer();
-    ```
-  </Tabs.Tab>
-</Tabs>
-
----
-
-## Streaming API (`/stream`)
-
-The `/stream` endpoint uses Server-Sent Events (SSE) to push data in real-time. This is ideal for applications where you want to display the response as it's being generated, such as in a live chatbot interface.
-
-### Request
-
--   **Endpoint:** `/stream`
--   **Method:** `POST`
--   **Payload:** Same as the non-streaming API.
-
-### Response (SSE Stream)
-
-The stream consists of multiple `data:` events, each containing a JSON object. Your client should listen for these events and process them based on their `type`.
-
-**Event Types:**
--   `answer`: A chunk of the agent's final answer.
--   `source`: A document or source used by the agent.
--   `thought`: A reasoning step from the agent (for ReAct agents).
--   `id`: The unique `conversation_id` for the interaction.
--   `error`: An error message.
--   `end`: A final message indicating the stream has concluded.
-
-### Examples
-
-<Tabs items={['cURL', 'Python', 'JavaScript']}>
-  <Tabs.Tab>
-    ```bash
-    curl -X POST http://localhost:7091/stream \
-    -H "Content-Type: application/json" \
-    -H "Accept: text/event-stream" \
-    -d '{
-        "question": "your question here",
-        "api_key": "your_agent_api_key"
-    }'
-    ```
-  </Tabs.Tab>
-  <Tabs.Tab>
-    ```python
-    import requests
-    import json
-
-    API_URL = "http://localhost:7091/stream"
-    payload = {
-        "question": "your question here",
-        "api_key": "your_agent_api_key"
-    }
-
-    with requests.post(API_URL, json=payload, stream=True) as r:
-        for line in r.iter_lines():
-            if line:
-                decoded_line = line.decode('utf-8')
-                if decoded_line.startswith('data: '):
-                    try:
-                        data = json.loads(decoded_line[6:])
-                        print(data)
-                    except json.JSONDecodeError:
-                        pass
-    ```
-  </Tabs.Tab>
-  <Tabs.Tab>
-    ```javascript
-    const apiUrl = 'http://localhost:7091/stream';
-    const apiKey = 'your_agent_api_key';
-    const question = 'your question here';
-
-    async function getStream() {
-    try {
-        const response = await fetch(apiUrl, {
-        method: 'POST',
-        headers: {
-            'Content-Type': 'application/json',
-            'Accept': 'text/event-stream'
-        },
-        // Corrected line: 'apiKey' is changed to 'api_key'
-        body: JSON.stringify({ question, api_key: apiKey }),
-        });
-
-        if (!response.ok) {
-        throw new Error(`HTTP error! Status: ${response.status}`);
-        }
-
-        const reader = response.body.getReader();
-        const decoder = new TextDecoder();
-
-        while (true) {
-        const { done, value } = await reader.read();
-        if (done) break;
-
-        const chunk = decoder.decode(value, { stream: true });
-        // Note: This parsing method assumes each chunk contains whole lines.
-        // For a more robust production implementation, buffer the chunks
-        // and process them line by line.
-        const lines = chunk.split('\n');
-        
-        for (const line of lines) {
-            if (line.startsWith('data: ')) {
-            try {
-                const data = JSON.parse(line.substring(6));
-                console.log(data);
-            } catch (e) {
-                console.error("Failed to parse JSON from SSE event:", e);
-            }
-            }
-        }
-        }
-    } catch (error) {
-        console.error("Failed to fetch stream:", error);
-    }
-    }
-
-    getStream();
-    ```
-  </Tabs.Tab>
-</Tabs>

docs/pages/Agents/basics.mdx

@@ -1,109 +0,0 @@
----
-title: Understanding DocsGPT Agents
-description: Learn about DocsGPT Agents, their types, how to create and manage them, and how they can enhance your interaction with documents and tools.
----
-
-import { Callout } from 'nextra/components';
-import Image from 'next/image'; // Assuming you might want to embed images later, like the ones you uploaded.
-
-# Understanding DocsGPT Agents 🤖
-
-DocsGPT Agents are advanced, configurable AI entities designed to go beyond simple question-answering. They act as specialized assistants or workers that combine instructions (prompts), knowledge (document sources), and capabilities (tools) to perform a wide range of tasks, automate workflows, and provide tailored interactions.
-
-Think of an Agent as a pre-configured version of DocsGPT, fine-tuned for a specific purpose, such as classifying documents, responding to new form submissions, or validating emails.
-
-## Why Use Agents?
-
-* **Personalization:** Create AI assistants that behave and respond according to specific roles or personas.
-* **Task Specialization:** Design agents focused on particular tasks, like customer support, data extraction, or content generation.
-* **Knowledge Integration:** Equip agents with specific document sources, making them experts in particular domains.
-* **Tool Utilization:** Grant agents access to various tools, allowing them to interact with external services, fetch live data, or perform actions.
-* **Automation:** Automate repetitive tasks by defining an agent's behavior and integrating it via webhooks or other means.
-* **Shareability:** Share your custom-configured agents with others or use agents shared with you.
-
-Agents provide a more structured and powerful way to leverage LLMs compared to a standard chat interface, as they come with a pre-defined context, instruction set, and set of capabilities.
-
-## Core Components of an Agent
-
-When you create or configure an agent, you'll work with these key components:  
-
-**Meta:**
-    * **Agent Name:** A user-friendly name to identify the agent (e.g., "Support Ticket Classifier," "Product Spec Expert").
-    * **Describe your agent:** A brief description for you or users to understand the agent's purpose.
-
-**Source:**
-    * **Select source:** The knowledge base for the agent. You can select from previously uploaded documents or data sources. This is what the agent will "know."
-    * **Chunks per query:** A numerical value determining how many relevant text chunks from the selected source are sent to the LLM with each query. This helps manage context length and relevance.
-
-**Prompt:**
-The main set of instructions or system [prompt](/Guides/Customising-prompts) that defines the agent's persona, objectives, constraints, and how it should behave or respond.
-
-**Tools:** A selection of available [DocsGPT Tools](/Tools/basics) that the agent can use to perform actions or access external information.
-
-**Agent type:** The underlying operational logic or architecture the agent uses. DocsGPT supports different types of agents, each suited for different kinds of tasks.
-
-## Understanding Agent Types
-
-DocsGPT allows for different "types" of agents, each with a distinct way of processing information and generating responses. The code for these agent types can be found in the `application/agents/` directory.
-
-### 1. Classic Agent (`classic_agent.py`)
-
-**How it works:** The Classic Agent follows a traditional Retrieval Augmented Generation (RAG) approach.
-    1.  **Retrieve:** When a query is made, it first searches the selected Source documents for relevant information.
-    2.  **Augment:** This retrieved data is then added to the context, along with the main Prompt and the user's query.
-    3.  **Generate:** The LLM generates a response based on this augmented context. It can also utilize any configured tools if the LLM decides they are necessary.
-
-**Best for:**
-    * Direct question-answering over a specific set of documents.
-    * Tasks where the primary goal is to extract and synthesize information from the provided sources.
-    * Simpler tool integrations where the decision to use a tool is straightforward.
-
-### 2. ReAct Agent (`react_agent.py`)
-
-**How it works:** The ReAct Agent employs a more sophisticated "Reason and Act" framework. This involves a multi-step process:
-    1.  **Plan (Thought):** Based on the query, its prompt, and available tools/sources, the LLM first generates a plan or a sequence of thoughts on how to approach the problem. You might see this output as a "thought" process during generation.
-    2.  **Act:** The agent then executes actions based on this plan. This might involve querying its sources, using a tool, or performing internal reasoning.
-    3.  **Observe:** It gathers observations from the results of its actions (e.g., data from a tool, snippets from documents).
-    4.  **Repeat (if necessary):** Steps 2 and 3 can be repeated as the agent refines its approach or gathers more information.
-    5.  **Conclude:** Finally, it generates the final answer based on the initial query and all accumulated observations.
-
-**Best for:**
-    * More complex tasks that require multi-step reasoning or problem-solving.
-    * Scenarios where the agent needs to dynamically decide which tools to use and in what order, based on intermediate results.
-    * Interactive tasks where the agent needs to "think" through a problem.
-
-<Callout type="info">
-Developers looking to introduce new agent architectures can explore the `application/agents/` directory. `classic_agent.py` and `react_agent.py` serve as excellent starting points, demonstrating how to inherit from `BaseAgent` and structure agent logic.
-</Callout>
-
-## Navigating and Managing Agents in DocsGPT
-
-You can easily access and manage your agents through the DocsGPT user interface. Recently used agents appear at the top of the left sidebar for quick access. Below these, the "Manage Agents" button will take you to the main Agents page.
-
-### Creating a New Agent
-
-1.  Navigate to the "Agents" page.
-2.  Click the **"New Agent"** button.
-3.  You will be presented with the "New Agent" configuration screen:
-
-<Image
-  src="/new-agent.png"
-  alt="API Tool configuration example for phone validation"
-  width={800}
-  height={450}
-  style={{ margin: '1em auto', display: 'block', borderRadius: '8px' }}
-/>
-
-4.  Fill in the fields as described in the "Core Components of an Agent" section.
-5.  Once configured, you can **"Save Draft"** to continue editing later or **"Publish"** to make the agent active.
-
-## Interacting with and Editing Agents
-
-Once an agent is created, you can:
-
-* **Chat with it:** Select the agent to start an interaction.
-* **View Logs:** Access usage statistics, monitor token consumption per interaction, and review user message feedbacks. This is crucial for understanding how your agent is being used and performing.
-* **Edit an Agent:**
-    * Modify any of its configuration settings (name, description, source, prompt, tools, type).
-    * **Generate a Public Link:** From the edit screen, you can create a shareable public link that allows others to import and use your agent.
-    * **Get a Webhook URL:** You can also obtain a Webhook URL for the agent. This allows external applications or services to trigger the agent and receive responses programmatically, enabling powerful integrations and automations.

docs/pages/Agents/webhooks.mdx

@@ -1,152 +0,0 @@
----
-title: Triggering Agents with Webhooks
-description: Learn how to automate and integrate DocsGPT Agents using webhooks for asynchronous task execution.
----
-
-import { Callout, Tabs } from 'nextra/components';
-
-# Triggering Agents with Webhooks
-
-Agent Webhooks provide a powerful mechanism to trigger an agent's execution from external systems. Unlike the direct API which provides an immediate response, webhooks are designed for **asynchronous** operations. When you call a webhook, DocsGPT enqueues the agent's task for background processing and immediately returns a `task_id`. You then use this ID to poll for the result.
-
-This workflow is ideal for integrating with services that expect a quick initial response (e.g., form submissions) or for triggering long-running tasks without tying up a client connection.
-
-Each agent has its own unique webhook URL, which can be generated from the agent's edit page in the DocsGPT UI. This URL includes a secure token for authentication.
-
-### API Endpoints
-
-- **Webhook URL:** `http://localhost:7091/api/webhooks/agents/{AGENT_WEBHOOK_TOKEN}`
-- **Task Status URL:** `http://localhost:7091/api/task_status`
-
-<Callout type="info">
-For DocsGPT Cloud, use `https://gptcloud.arc53.com/` as the base URL.
-</Callout>
-
-For more technical details, you can explore the API swagger documentation available for the cloud version or your local instance.
-
----
-
-## The Webhook Workflow
-
-The process involves two main steps: triggering the task and polling for the result.
-
-### Step 1: Trigger the Webhook
-
-Send an HTTP `POST` request to the agent's unique webhook URL with the required payload. The structure of this payload should match what the agent's prompt and tools are designed to handle.
-
--   **Method:** `POST`
--   **Response:** A JSON object with a `task_id`. `{"task_id": "a1b2c3d4-e5f6-..."}`
-
-<Tabs items={['cURL', 'Python', 'JavaScript']}>
-  <Tabs.Tab>
-    ```bash
-    curl -X POST \
-      http://localhost:7091/api/webhooks/agents/your_webhook_token \
-      -H "Content-Type: application/json" \
-      -d '{"question": "Your message to agent"}'
-    ```
-  </Tabs.Tab>
-  <Tabs.Tab>
-    ```python
-    import requests
-
-    WEBHOOK_URL = "http://localhost:7091/api/webhooks/agents/your_webhook_token"
-    payload = {"question": "Your message to agent"}
-
-    try:
-        response = requests.post(WEBHOOK_URL, json=payload)
-        response.raise_for_status()
-        task_id = response.json().get("task_id")
-        print(f"Task successfully created with ID: {task_id}")
-    except requests.exceptions.RequestException as e:
-        print(f"Error triggering webhook: {e}")
-    ```
-  </Tabs.Tab>
-  <Tabs.Tab>
-    ```javascript
-    const webhookUrl = 'http://localhost:7091/api/webhooks/agents/your_webhook_token';
-    const payload = { question: 'Your message to agent' };
-
-    async function triggerWebhook() {
-      try {
-        const response = await fetch(webhookUrl, {
-          method: 'POST',
-          headers: { 'Content-Type': 'application/json' },
-          body: JSON.stringify(payload)
-        });
-        if (!response.ok) throw new Error(`HTTP error! ${response.status}`);
-        const data = await response.json();
-        console.log(`Task successfully created with ID: ${data.task_id}`);
-        return data.task_id;
-      } catch (error) {
-        console.error('Error triggering webhook:', error);
-      }
-    }
-    
-    triggerWebhook();
-    ```
-  </Tabs.Tab>
-</Tabs>
-
-### Step 2: Poll for the Result
-
-Once you have the `task_id`, periodically send a `GET` request to the `/api/task_status` endpoint until the task `status` is `SUCCESS` or `FAILURE`.
-
-- **`status`**: The current state of the task (`PENDING`, `STARTED`, `SUCCESS`, `FAILURE`).
-- **`result`**: The final output from the agent, available when the status is `SUCCESS` or `FAILURE`.
-
-<Tabs items={['cURL', 'Python', 'JavaScript']}>
-  <Tabs.Tab>
-    ```bash
-    # Replace the task_id with the one you received
-    curl http://localhost:7091/api/task_status?task_id=YOUR_TASK_ID
-    ```
-  </Tabs.Tab>
-  <Tabs.Tab>
-    ```python
-    import requests
-    import time
-
-    STATUS_URL = "http://localhost:7091/api/task_status"
-    task_id = "YOUR_TASK_ID"
-
-    while True:
-        response = requests.get(STATUS_URL, params={"task_id": task_id})
-        data = response.json()
-        status = data.get("status")
-        print(f"Current task status: {status}")
-
-        if status in ["SUCCESS", "FAILURE"]:
-            print("Final Result:")
-            print(data.get("result"))
-            break
-        
-        time.sleep(2)
-    ```
-  </Tabs.Tab>
-  <Tabs.Tab>
-    ```javascript
-    const statusUrl = 'http://localhost:7091/api/task_status';
-    const taskId = 'YOUR_TASK_ID';
-
-    const sleep = (ms) => new Promise(resolve => setTimeout(resolve, ms));
-
-    async function pollForResult() {
-      while (true) {
-        const response = await fetch(`${statusUrl}?task_id=${taskId}`);
-        const data = await response.json();
-        const status = data.status;
-        console.log(`Current task status: ${status}`);
-
-        if (status === 'SUCCESS' || status === 'FAILURE') {
-          console.log('Final Result:', data.result);
-          break;
-        }
-        await sleep(2000);
-      }
-    }
-
-    pollForResult();
-    ```
-  </Tabs.Tab>
-</Tabs>

docs/pages/Deploying/Docker-Deploying.mdx

@@ -37,7 +37,7 @@ The fastest way to try out DocsGPT is by using the public API endpoint. This req
     Open the `.env` file and add the following lines:
 
     ```
-    LLM_PROVIDER=docsgpt
+    LLM_NAME=docsgpt
     VITE_API_STREAMING=true
     ```
 
@@ -93,16 +93,16 @@ There are two Ollama optional files:
 
 3.  **Pull the Ollama Model:**
 
-    **Crucially, after launching with Ollama, you need to pull the desired model into the Ollama container.**  Find the `LLM_NAME` you configured in your `.env` file (e.g., `llama3.2:1b`). Then execute the following command to pull the model *inside* the running Ollama container:
+    **Crucially, after launching with Ollama, you need to pull the desired model into the Ollama container.**  Find the `MODEL_NAME` you configured in your `.env` file (e.g., `llama3.2:1b`). Then execute the following command to pull the model *inside* the running Ollama container:
 
     ```bash
-    docker compose -f deployment/docker-compose.yaml -f deployment/optional/docker-compose.optional.ollama-cpu.yaml exec -it ollama ollama pull <LLM_NAME>
+    docker compose -f deployment/docker-compose.yaml -f deployment/optional/docker-compose.optional.ollama-cpu.yaml exec -it ollama ollama pull <MODEL_NAME>
     ```
     or (for GPU):
      ```bash
-    docker compose -f deployment/docker-compose.yaml -f deployment/optional/docker-compose.optional.ollama-gpu.yaml exec -it ollama ollama pull <LLM_NAME>
+    docker compose -f deployment/docker-compose.yaml -f deployment/optional/docker-compose.optional.ollama-gpu.yaml exec -it ollama ollama pull <MODEL_NAME>
     ```
-    Replace `<LLM_NAME>` with the actual model name from your `.env` file.
+    Replace `<MODEL_NAME>` with the actual model name from your `.env` file.
 
 4.  **Access DocsGPT in your browser:**
 

docs/pages/Deploying/DocsGPT-Settings.mdx

@@ -20,9 +20,9 @@ The easiest and recommended way to configure basic settings is by using a `.env`
 **Example `.env` file structure:**
 
 ```
-LLM_PROVIDER=openai
+LLM_NAME=openai
 API_KEY=YOUR_OPENAI_API_KEY
-LLM_NAME=gpt-4o
+MODEL_NAME=gpt-4o
 ```
 
 ### 2. Configuration via `settings.py` file (Advanced)
@@ -37,7 +37,7 @@ While modifying `settings.py` offers more flexibility, it's generally recommende
 
 Here are some of the most fundamental settings you'll likely want to configure:
 
-- **`LLM_PROVIDER`**: This setting determines which Large Language Model (LLM) provider DocsGPT will use.  It tells DocsGPT which API to interact with.
+- **`LLM_NAME`**: This setting determines which Large Language Model (LLM) provider DocsGPT will use.  It tells DocsGPT which API to interact with.
 
     - **Common values:**
         - `docsgpt`:  Use the DocsGPT Public API Endpoint (simple and free, as offered in `setup.sh` option 1).
@@ -49,11 +49,11 @@ Here are some of the most fundamental settings you'll likely want to configure:
         - `azure_openai`: Use Azure OpenAI Service.
         - `openai` (when using local inference engines like Ollama, Llama.cpp, TGI, etc.):  This signals DocsGPT to use an OpenAI-compatible API format, even if the actual LLM is running locally.
 
-- **`LLM_NAME`**:  Specifies the specific model to use from the chosen LLM provider. The available models depend on the `LLM_PROVIDER` you've selected.
+- **`MODEL_NAME`**:  Specifies the specific model to use from the chosen LLM provider. The available models depend on the `LLM_NAME` you've selected.
 
     - **Examples:**
-        - For `LLM_PROVIDER=openai`: `gpt-4o`
-        - For `LLM_PROVIDER=google`: `gemini-2.0-flash`
+        - For `LLM_NAME=openai`: `gpt-4o`
+        - For `LLM_NAME=google`: `gemini-2.0-flash`
         - For local models (e.g., Ollama): `llama3.2:1b` (or any model name available in your setup).
 
 - **`EMBEDDINGS_NAME`**:  This setting defines which embedding model DocsGPT will use to generate vector embeddings for your documents. Embeddings are numerical representations of text that allow DocsGPT to understand the semantic meaning of your documents for efficient search and retrieval.
@@ -63,7 +63,7 @@ Here are some of the most fundamental settings you'll likely want to configure:
 
 - **`API_KEY`**:  Required for most cloud-based LLM providers.  This is your authentication key to access the LLM provider's API. You'll need to obtain this key from your chosen provider's platform.
 
-- **`OPENAI_BASE_URL`**:  Specifically used when `LLM_PROVIDER` is set to `openai` but you are connecting to a local inference engine (like Ollama, Llama.cpp, etc.) that exposes an OpenAI-compatible API.  This setting tells DocsGPT where to find your local LLM server.
+- **`OPENAI_BASE_URL`**:  Specifically used when `LLM_NAME` is set to `openai` but you are connecting to a local inference engine (like Ollama, Llama.cpp, etc.) that exposes an OpenAI-compatible API.  This setting tells DocsGPT where to find your local LLM server.
 
 ## Configuration Examples
 
@@ -74,9 +74,9 @@ Let's look at some concrete examples of how to configure these settings in your
 To use OpenAI's `gpt-4o` model, you would configure your `.env` file like this:
 
 ```
-LLM_PROVIDER=openai
+LLM_NAME=openai
 API_KEY=YOUR_OPENAI_API_KEY  # Replace with your actual OpenAI API key
-LLM_NAME=gpt-4o
+MODEL_NAME=gpt-4o
 ```
 
 Make sure to replace `YOUR_OPENAI_API_KEY` with your actual OpenAI API key.
@@ -86,57 +86,14 @@ Make sure to replace `YOUR_OPENAI_API_KEY` with your actual OpenAI API key.
 To use a local Ollama server with the `llama3.2:1b` model, you would configure your `.env` file like this:
 
 ```
-LLM_PROVIDER=openai # Using OpenAI compatible API format for local models
+LLM_NAME=openai # Using OpenAI compatible API format for local models
 API_KEY=None      # API Key is not needed for local Ollama
-LLM_NAME=llama3.2:1b
+MODEL_NAME=llama3.2:1b
 OPENAI_BASE_URL=http://host.docker.internal:11434/v1 # Default Ollama API URL within Docker
 EMBEDDINGS_NAME=huggingface_sentence-transformers/all-mpnet-base-v2 # You can also run embeddings locally if needed
 ```
 
-In this case, even though you are using Ollama locally, `LLM_PROVIDER` is set to `openai` because Ollama (and many other local inference engines) are designed to be API-compatible with OpenAI.  `OPENAI_BASE_URL` points DocsGPT to the local Ollama server.
-
-## Authentication Settings
-
-DocsGPT includes a JWT (JSON Web Token) based authentication feature for managing sessions or securing local deployments while allowing access.
-
-- **`AUTH_TYPE`**: This setting in your `.env` file or `settings.py` determines the authentication method.
-    
-    - **Possible values:**
-        - `None` (or not set): No authentication is used.
-        - `simple_jwt`: A single, long-lived JWT token is generated and used for all authenticated requests. This is useful for securing a local deployment with a shared secret.
-        - `session_jwt`: Unique JWT tokens are generated for sessions, typically for individual users or temporary access.
-    - If `AUTH_TYPE` is set to `simple_jwt` or `session_jwt`, then a `JWT_SECRET_KEY` is required.
-- **`JWT_SECRET_KEY`**: This is a crucial secret key used to sign and verify JWTs.
-    
-    - It can be set directly in your `.env` file or `settings.py`.
-    - **Automatic Key Generation**: If `AUTH_TYPE` is `simple_jwt` or `session_jwt` and `JWT_SECRET_KEY` is _not_ set in your environment variables or `settings.py`, DocsGPT will attempt to:
-        1. Read the key from a file named `.jwt_secret_key` in the project's root directory.
-        2. If the file doesn't exist, it will generate a new 32-byte random key, save it to `.jwt_secret_key`, and use it for the session. This ensures that the key persists across application restarts.
-    - **Security Note**: It's vital to keep this key secure. If you set it manually, choose a strong, random string.
-
-**How it works:**
-
-- When `AUTH_TYPE` is set to `simple_jwt`, a token is generated at startup (if not already present or configured) and printed to the console. This token should be included in the `Authorization` header of your API requests as a Bearer token (e.g., `Authorization: Bearer YOUR_SIMPLE_JWT_TOKEN`).
-- When `AUTH_TYPE` is set to `session_jwt`:
-    - Clients can request a new token from the `/api/generate_token` endpoint.
-    - This token should then be included in the `Authorization` header for subsequent requests.
-- The backend verifies the JWT token provided in the `Authorization` header for protected routes.
-- The `/api/config` endpoint can be used to check the current `auth_type` and whether authentication is required.
-
-**Frontend Token Input for `simple_jwt`:**
-
-<img 
-  src="/jwt-input.png" 
-  alt="Frontend prompt for JWT Token" 
-  style={{ 
-    width: '500px', 
-    maxWidth: '100%', 
-    display: 'block', 
-    margin: '1em auto' 
-  }} 
-/>
-
-If you have configured `AUTH_TYPE=simple_jwt`, the DocsGPT frontend will prompt you to enter the JWT token if it's not already set or is invalid. You'll need to paste the `SIMPLE_JWT_TOKEN` (which is printed to your console when the backend starts) into this field to access the application.
+In this case, even though you are using Ollama locally, `LLM_NAME` is set to `openai` because Ollama (and many other local inference engines) are designed to be API-compatible with OpenAI.  `OPENAI_BASE_URL` points DocsGPT to the local Ollama server.
 
 ## Exploring More Settings
 

docs/pages/Extensions/chat-widget copy.md

@@ -0,0 +1,212 @@
+# Setting up the DocsGPT Widget in Your React Project
+
+## Introduction:
+The DocsGPT Widget is a powerful tool that allows you to integrate AI-powered documentation assistance into your web applications. This guide will walk you through the installation and usage of the DocsGPT Widget in your React project. Whether you're building a web app or a knowledge base, this widget can enhance your user experience.
+
+## Installation
+First, make sure you have Node.js and npm installed in your project. Then go to your project and install a new dependency: `npm install docsgpt`.
+
+## Usage
+In the file where you want to use the widget, import it and include the CSS file:
+```js
+import { DocsGPTWidget } from "docsgpt";
+```
+
+
+Now, you can use the widget in your component like this :
+```jsx
+<DocsGPTWidget
+  apiHost="https://your-docsgpt-api.com"
+  apiKey=""
+  avatar = "https://d3dg1063dc54p9.cloudfront.net/cute-docsgpt.png"
+  title = "Get AI assistance"
+  description = "DocsGPT's AI Chatbot is here to help"
+  heroTitle = "Welcome to DocsGPT !"
+  heroDescription="This chatbot is built with DocsGPT and utilises GenAI, 
+  please review important information using sources."
+  theme = "dark"
+  buttonIcon = "https://your-icon"
+  buttonBg = "#222327"
+/>
+```
+## Props Table for DocsGPT Widget
+
+| **Prop**          | **Type**         | **Default Value**                                           | **Description**                                                                                     |
+|--------------------|------------------|-------------------------------------------------------------|-----------------------------------------------------------------------------------------------------|
+| **`apiHost`**      | `string`         | `"https://gptcloud.arc53.com"`                                     | The URL of your DocsGPT API for vector search and chatbot queries.                                 |
+| **`apiKey`**       | `string`         | `""`                                                        | Your API key for authentication. Can be left empty if authentication is not required.              |
+| **`avatar`**       | `string`         | `"https://d3dg1063dc54p9.cloudfront.net/cute-docsgpt.png"`   | Specifies the URL of the avatar or image representing the chatbot.                                 |
+| **`title`**        | `string`         | `"Get AI assistance"`                                       | Sets the title text displayed in the chatbot interface.                                            |
+| **`description`**  | `string`         | `"DocsGPT's AI Chatbot is here to help"`                     | Provides a brief description of the chatbot's purpose or functionality.                            |
+| **`heroTitle`**    | `string`         | `"Welcome to DocsGPT !"`                                    | Displays a welcome title when users interact with the chatbot.                                     |
+| **`heroDescription`** | `string`     | `"This chatbot is built with DocsGPT and utilises GenAI, please review important information using sources."` | Provides additional introductory text or information about the chatbot's capabilities.             |
+| **`theme`**        | `"dark" \| "light"` | `"dark"`                                                  | Allows you to select the theme for the chatbot interface. Accepts `"dark"` or `"light"`.           |
+| **`buttonIcon`**   | `string`         | `"https://your-icon"`                                        | Specifies the URL of the icon image for the widget's launch button.                                |
+| **`buttonBg`**     | `string`         | `"#222327"`                                                 | Sets the background color of the widget's launch button.                                           |
+| **`size`**         | `"small" \| "medium"` | `"medium"`                                              | Sets the size of the widget. Options are `"small"` or `"medium"`.                                  |
+
+---
+
+## Notes
+- **Customizing Props:** All properties can be overridden when embedding the widget. For example, you can provide a unique avatar, title, or color scheme to better align with your brand.
+- **Default Theme:** The widget defaults to the dark theme unless explicitly set to `"light"`. 
+- **API Key:** If the `apiKey` is not required for your application, leave it empty.
+
+This table provides a clear overview of the customization options available for tailoring the DocsGPT widget to fit your application.
+
+
+## How to use DocsGPTWidget with [Nextra](https://nextra.site/) (Next.js + MDX)
+Install your widget as described above and then go to your `pages/` folder and create a new file `_app.js` with the following content:
+```js
+import { DocsGPTWidget } from "docsgpt";
+
+export default function MyApp({ Component, pageProps }) {
+    return (
+        <>
+            <Component {...pageProps} />
+            <DocsGPTWidget selectDocs="local/docsgpt-sep.zip/"/>
+        </>
+    )
+}
+```  
+## How to use DocsGPTWidget with HTML
+```html
+<!DOCTYPE html>
+<html lang="en">
+  <head>
+    <meta charset="UTF-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <meta http-equiv="X-UA-Compatible" content="ie=edge" />
+    <title>HTML + CSS</title>
+    <link rel="stylesheet" href="styles.css" />
+  </head>
+  <body>
+    <h1>This is a simple HTML + CSS template!</h1>
+    <div id="app"></div>
+    <!-- Include the widget script from dist/modern or dist/legacy -->
+    <script
+      src="https://unpkg.com/docsgpt/dist/modern/main.js"
+      type="module"
+    ></script>
+    <script type="module">
+      window.onload = function () {
+        renderDocsGPTWidget("app", {
+          apiKey: "",
+          size: "medium",
+        });
+      };
+    </script>
+  </body>
+</html>
+```
+To link the widget to your api and your documents you can pass parameters to the renderDocsGPTWidget('div id', { parameters }).
+```html
+<!DOCTYPE html>
+<html lang="en">
+<head>
+  <meta charset="UTF-8" />
+  <meta http-equiv="X-UA-Compatible" content="IE=edge" />
+  <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+  <title>DocsGPT Widget</title>
+  <script src="https://unpkg.com/docsgpt/dist/modern/main.js" type="module"></script>  
+</head>
+<body>
+  <div id="app"></div>
+  <!-- Include the widget script from dist/modern or dist/legacy -->
+  <script type="module">
+    window.onload = function() {
+      renderDocsGPTWidget('app', {
+        apiHost: 'http://localhost:7001',
+        apiKey:"",
+        avatar: 'https://d3dg1063dc54p9.cloudfront.net/cute-docsgpt.png',
+        title: 'Get AI assistance',
+        description: "DocsGPT's AI Chatbot is here to help",
+        heroTitle: 'Welcome to DocsGPT!',
+        heroDescription: 'This chatbot is built with DocsGPT and utilises GenAI, please review important information using sources.',
+        theme:"dark",
+        buttonIcon:"https://your-icon",
+        buttonBg:"#222327"
+      });
+    }
+  </script>
+</body>
+</html>
+```
+
+# SearchBar
+
+The `SearchBar` component is an interactive search bar designed to provide search results based on **vector similarity search**. It also includes the capability to open the AI Chatbot, enabling users to query.
+
+---
+
+### Importing the Component
+```tsx
+import { SearchBar } from "docsgpt-react";
+```
+
+---
+
+### Usage Example
+```tsx
+<SearchBar 
+    apiKey="your-api-key"
+    apiHost="https://gptcloud.arc53.com"
+    theme="light"
+    placeholder="Search or Ask AI..."
+    width="300px"
+/>
+```
+
+---
+
+## HTML embedding for Search bar
+
+```html
+<!DOCTYPE html>
+<html lang="en">
+<head>
+  <meta charset="UTF-8">
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  <title>SearchBar Embedding</title>
+  <script src="https://unpkg.com/docsgpt/dist/modern/main.js"></script> <!-- The bundled JavaScript file -->
+</head>
+<body>
+  <!-- Element where the SearchBar will render -->
+  <div id="search-bar-container"></div>
+
+  <script>
+    // Render the SearchBar into the specified element
+    renderSearchBar('search-bar-container', {
+      apiKey: 'your-api-key-here',
+      apiHost: 'https://your-api-host.com',
+      theme: 'light',
+      placeholder: 'Search here...',
+      width: '300px'
+    });
+  </script>
+</body>
+</html>
+
+```
+
+### Props
+
+| **Prop**       | **Type**  | **Default Value**                   | **Description**                                                                                  |
+|-----------------|-----------|-------------------------------------|--------------------------------------------------------------------------------------------------|
+| **`apiKey`**    | `string`  | `"74039c6d-bff7-44ce-ae55-2973cbf13837"` | Your API key generated from the app. Used for authenticating requests.                         |
+| **`apiHost`**   | `string`  | `"https://gptcloud.arc53.com"`       | The base URL of the server hosting the vector similarity search and chatbot services.           |
+| **`theme`**     | `"dark" \| "light"` | `"dark"`                            | The theme of the search bar. Accepts `"dark"` or `"light"`.                                     |
+| **`placeholder`** | `string` | `"Search or Ask AI..."`             | Placeholder text displayed in the search input field.                                           |
+| **`width`**     | `string`  | `"256px"`                          | Width of the search bar. Accepts any valid CSS width value (e.g., `"300px"`, `"100%"`, `"20rem"`). |
+
+
+Feel free to reach out if you need help customizing or extending the `SearchBar`!
+
+## Our github
+
+[DocsGPT](https://github.com/arc53/DocsGPT)
+
+You can find the source code in the extensions/react-widget folder.
+
+For more information about React, refer to this [link here](https://react.dev/learn)
+

docs/pages/Guides/How-to-use-different-LLM.mdx

@@ -32,9 +32,9 @@ Choose the LLM of your choice.
 ### For Open source llm change:
 <Steps>
 ### Step 1
-For open source version please edit `LLM_PROVIDER`, `LLM_NAME` and others in the .env file. Refer to [⚙️ App Configuration](/Deploying/DocsGPT-Settings) for more information.
+For open source version please edit `LLM_NAME`, `MODEL_NAME` and others in the .env file. Refer to [⚙️ App Configuration](/Deploying/DocsGPT-Settings) for more information.
 ### Step 2
-Visit [☁️ Cloud Providers](/Models/cloud-providers) for the updated list of online models. Make sure you have the right API_KEY and correct LLM_PROVIDER.
+Visit [☁️ Cloud Providers](/Models/cloud-providers) for the updated list of online models. Make sure you have the right API_KEY and correct LLM_NAME.
 For self-hosted please visit [🖥️ Local Inference](/Models/local-inference). 
 </Steps>
 

docs/pages/Models/cloud-providers.mdx

@@ -13,15 +13,15 @@ The primary method for configuring your LLM provider in DocsGPT is through the `
 
 To connect to a cloud LLM provider, you will typically need to configure the following basic settings in your `.env` file:
 
-*   **`LLM_PROVIDER`**:  This setting is essential and identifies the specific cloud provider you wish to use (e.g., `openai`, `google`, `anthropic`).
-*   **`LLM_NAME`**:  Specifies the exact model you want to utilize from your chosen provider (e.g., `gpt-4o`, `gemini-2.0-flash`, `claude-3-5-sonnet-latest`).  Refer to your provider's documentation for a list of available models.
+*   **`LLM_NAME`**:  This setting is essential and identifies the specific cloud provider you wish to use (e.g., `openai`, `google`, `anthropic`).
+*   **`MODEL_NAME`**:  Specifies the exact model you want to utilize from your chosen provider (e.g., `gpt-4o`, `gemini-2.0-flash`, `claude-3-5-sonnet-latest`).  Refer to your provider's documentation for a list of available models.
 *   **`API_KEY`**:  Almost all cloud LLM providers require an API key for authentication. Obtain your API key from your chosen provider's platform and securely store it in your `.env` file.
 
 ## Explicitly Supported Cloud Providers
 
-DocsGPT offers direct, streamlined support for the following cloud LLM providers, making configuration straightforward.  The table below outlines the `LLM_PROVIDER` and example `LLM_NAME` values to use for each provider in your `.env` file.
+DocsGPT offers direct, streamlined support for the following cloud LLM providers, making configuration straightforward.  The table below outlines the `LLM_NAME` and example `MODEL_NAME` values to use for each provider in your `.env` file.
 
-| Provider                     | `LLM_PROVIDER` | Example `LLM_NAME`          |
+| Provider                     | `LLM_NAME`     | Example `MODEL_NAME`        |
 | :--------------------------- | :------------- | :-------------------------- |
 | DocsGPT Public API           | `docsgpt`      | `None`                      |
 | OpenAI                       | `openai`       | `gpt-4o`                    |
@@ -35,16 +35,16 @@ DocsGPT offers direct, streamlined support for the following cloud LLM providers
 
 DocsGPT's flexible architecture allows you to connect to any cloud provider that offers an API compatible with the OpenAI API standard. This opens up a vast ecosystem of LLM services.
 
-To connect to an OpenAI-compatible cloud provider, you will still use `LLM_PROVIDER=openai` in your `.env` file.  However, you will also need to specify the API endpoint of your chosen provider using the `OPENAI_BASE_URL` setting.  You will also likely need to provide an `API_KEY` and `LLM_NAME` as required by that provider.
+To connect to an OpenAI-compatible cloud provider, you will still use `LLM_NAME=openai` in your `.env` file.  However, you will also need to specify the API endpoint of your chosen provider using the `OPENAI_BASE_URL` setting.  You will also likely need to provide an `API_KEY` and `MODEL_NAME` as required by that provider.
 
 **Example for DeepSeek (OpenAI-Compatible API):**
 
 To connect to DeepSeek, which offers an OpenAI-compatible API, your `.env` file could be configured as follows:
 
 ```
-LLM_PROVIDER=openai
+LLM_NAME=openai
 API_KEY=YOUR_API_KEY # Your DeepSeek API key
-LLM_NAME=deepseek-chat # Or your desired DeepSeek model name
+MODEL_NAME=deepseek-chat # Or your desired DeepSeek model name
 OPENAI_BASE_URL=https://api.deepseek.com/v1 # DeepSeek's OpenAI API URL
 ```
 

docs/pages/Models/embeddings.md

@@ -60,7 +60,7 @@ To use OpenAI's `text-embedding-ada-002` embedding model, you need to set `EMBED
 **Example `.env` configuration for OpenAI Embeddings:**
 
 ```
-LLM_PROVIDER=openai
+LLM_NAME=openai
 API_KEY=YOUR_OPENAI_API_KEY # Your OpenAI API Key
 EMBEDDINGS_NAME=openai_text-embedding-ada-002
 ```

docs/pages/Models/local-inference.mdx

@@ -15,8 +15,8 @@ Setting up a local inference engine with DocsGPT is configured through environme
 
 To connect to a local inference engine, you will generally need to configure these settings in your `.env` file:
 
-*   **`LLM_PROVIDER`**:  Crucially set this to `openai`. This tells DocsGPT to use the OpenAI-compatible API format for communication, even though the LLM is local.
-*   **`LLM_NAME`**:  Specify the model name as recognized by your local inference engine. This might be a model identifier or left as `None` if the engine doesn't require explicit model naming in the API request.
+*   **`LLM_NAME`**:  Crucially set this to `openai`. This tells DocsGPT to use the OpenAI-compatible API format for communication, even though the LLM is local.
+*   **`MODEL_NAME`**:  Specify the model name as recognized by your local inference engine. This might be a model identifier or left as `None` if the engine doesn't require explicit model naming in the API request.
 *   **`OPENAI_BASE_URL`**:  This is essential. Set this to the base URL of your local inference engine's API endpoint. This tells DocsGPT where to find your local LLM server.
 *   **`API_KEY`**:  Generally, for local inference engines, you can set `API_KEY=None` as authentication is usually not required in local setups.
 
@@ -24,16 +24,16 @@ To connect to a local inference engine, you will generally need to configure the
 
 DocsGPT is readily configurable to work with the following local inference engines, all communicating via the OpenAI API format. Here are example `OPENAI_BASE_URL` values for each, based on default setups:
 
-| Inference Engine              | `LLM_PROVIDER` | `OPENAI_BASE_URL`          |
-| :---------------------------- | :------------- | :------------------------- |
-| LLaMa.cpp                     | `openai`       | `http://localhost:8000/v1`   |
-| Ollama                        | `openai`       | `http://localhost:11434/v1`  |
-| Text Generation Inference (TGI)| `openai`      | `http://localhost:8080/v1`   |
-| SGLang                        | `openai`       | `http://localhost:30000/v1`  |
-| vLLM                          | `openai`       | `http://localhost:8000/v1`   |
-| Aphrodite                     | `openai`       | `http://localhost:2242/v1`   |
-| FriendliAI                    | `openai`       | `http://localhost:8997/v1`   |
-| LMDeploy                      | `openai`       | `http://localhost:23333/v1` |
+| Inference Engine              | `LLM_NAME` | `OPENAI_BASE_URL`          |
+| :---------------------------- | :--------- | :------------------------- |
+| LLaMa.cpp                     | `openai`   | `http://localhost:8000/v1`   |
+| Ollama                        | `openai`   | `http://localhost:11434/v1`  |
+| Text Generation Inference (TGI)| `openai`   | `http://localhost:8080/v1`   |
+| SGLang                        | `openai`   | `http://localhost:30000/v1`  |
+| vLLM                          | `openai`   | `http://localhost:8000/v1`   |
+| Aphrodite                     | `openai`   | `http://localhost:2242/v1`   |
+| FriendliAI                    | `openai`   | `http://localhost:8997/v1`   |
+| LMDeploy                      | `openai`   | `http://localhost:23333/v1` |
 
 **Important Note on `localhost` vs `host.docker.internal`:**
 

docs/pages/Tools/_meta.json

@@ -1,14 +0,0 @@
-{
-  "basics": {
-    "title": "🔧 Tools Basics",
-    "href": "/Tools/basics"
-  },
-  "api-tool": {
-    "title": "🗝️ API Tool",
-    "href": "/Tools/api-tool"
-  },
-  "creating-a-tool": {
-    "title": "🛠️ Creating a Custom Tool",
-    "href": "/Tools/creating-a-tool"
-  }
-}

docs/pages/Tools/api-tool.mdx

@@ -1,153 +0,0 @@
----
-title: 🗝️ Generic API Tool
-description: Learn how to configure and use the API Tool in DocsGPT to connect with any RESTful API without writing custom code.
----
-
-import { Callout } from 'nextra/components';
-import Image from 'next/image';
-
-# Using the Generic API Tool
-
-The API Tool provides a no-code/low-code solution to make DocsGPT interact with third-party or internal RESTful APIs. It acts as a bridge, allowing the Large Language Model (LLM) to leverage external services based on your chat interactions.
- This guide will walk you through its capabilities, configuration, and best practices.
-
-## Introduction to the Generic API Tool
-
-**When to Use It:**
-    * Ideal for quickly integrating existing APIs where the interaction involves standard HTTP requests (GET, POST, PUT, DELETE).
-    * Suitable for fetching data to enrich answers (e.g., current weather, stock prices, product details).
-    * Useful for triggering simple actions in other systems (e.g., sending a notification, creating a basic task).
-
-**Contrast with Custom Python Tools:**
-    * **API Tool:** Best for straightforward API calls. Configuration is done through the DocsGPT UI.
-    * **Custom Python Tools:** Preferable when you need complex logic before or after the API call, handle non-standard authentication (like complex OAuth flows), manage multi-step API interactions, or require intricate data processing not easily managed by the LLM alone. See [Creating a Custom Tool](/Tools/creating-a-tool) for more.
-
-## Capabilities of the API Tool
-
-**Supported HTTP Methods:** You can configure actions using standard HTTP methods such as:
-    * `GET`: To retrieve data.
-    * `POST`: To submit data to create a new resource.
-    * `PUT`: To update an existing resource.
-    * `DELETE`: To remove a resource.
-
-**Request Configuration:**
-    * **Headers:** Define static or dynamic HTTP headers for authentication (e.g., API keys), content type specification, etc.
-    * **Query Parameters:** Specify URL query parameters, which can be static or dynamically filled by the LLM based on user input.
-    * **Request Body:** Define the structure of the request body (e.g., JSON), with fields that can be static or dynamically populated by the LLM.
-
-**Response Handling:**
-    * The API Tool executes the request and receives the raw response from the API (typically JSON or plain text).
-    * This raw response is then passed back to the LLM.
-    * The LLM uses this response, along with the context of your query and the description of the API tool action, to formulate an answer or decide on follow-up actions. The API tool itself doesn't deeply parse or transform the response beyond basic content type detection (e.g., loading JSON into a parsable object).
-
-## Configuring an API as a Tool
-
-You can configure the API Tool through the DocsGPT user interface, found in **Settings -> Tools**. When you add or modify an API Tool, you'll define specific actions that DocsGPT can perform.
-
-<Callout type="info">
-  The configuration involves defining how DocsGPT should call an API endpoint. Each configured API call essentially becomes a distinct "action" the LLM can choose to use.
-</Callout>
-
-Below is an example of how you might configure an API action, inspired by setting up a phone number validation service:
-
-<Image
-  src="/toolIcons/api-tool-example.png"
-  alt="API Tool configuration example for phone validation"
-  width={800}
-  height={450}
-  style={{ margin: '1em auto', display: 'block', borderRadius: '8px' }}
-/>
-_Figure 1: Example configuration for an API Tool action to validate phone numbers._
-
-**Defining an API Endpoint/Action:**
-
-When you configure a new API action, you'll fill in the following fields:
-
--   **`Name`:** A user-friendly name for this specific API action (e.g., "Phone-check" as in the image, or more specific like "ValidateUSPhoneNumber"). This helps in managing your tools.
--   **`Description`:** This is a **critical field**. Provide a clear and concise description of what the API action does, what kind of input it expects (implicitly), and what kind of output it provides. The LLM uses this description to understand when and how to use this action.
--   **`URL`:** The full endpoint URL for the API request.
--   **`HTTP Method`:** Select the appropriate HTTP method (e.g., GET, POST) from a dropdown.
--   **`Headers`:** You can add custom HTTP headers as key-value pairs (Name, Value). Indicate if the value should be `Filled by LLM` or is static. If filled by LLM, provide a `Description` for the LLM.
-
--   **`Query Parameters`:** For `GET` requests or when parameters are sent in the URL.
-    * **`Name`:** The name of the query parameter (e.g., `api_key`, `phone`).
-    * **`Type`:** The data type of the parameter (e.g., `string`).
-    * **`Filled by LLM` (Checkbox):**
-        - **Unchecked (Static):** The `Value` you provide will be used for every call (e.g., for an `api_key` that doesn't change).
-        - **Checked (Dynamic):** The LLM will extract the appropriate value from the user's chat query based on the `Description` you provide for this parameter. The `Value` field is typically left empty or contains a placeholder if `Filled by LLM` is checked.
-    * `Description`: Context for the LLM if the parameter is to be filled dynamically, or for your own reference if static.
-    * `Value`: The static value if not filled by LLM.
-
--   **`Request Body`:** Used to send data (commonly JSON) to the API. Similar to Query Parameters, you define fields with `Name`, `Type`, whether it's `Filled by LLM`, a `Description` for dynamic fields, and a static `Value` if applicable.
-
-**Response Handling Guidance for the LLM:**
-
-While the API Tool configuration UI doesn't have explicit fields for defining response parsing rules (like JSONPath extractors), you significantly influence how the LLM handles the response through:
-    * **Tool Action `Description`:** Clearly state what kind of information the API returns (e.g., "This API returns a JSON object with 'status' and 'location' fields for the phone number."). This helps the LLM know what to look for in the API's output.
-    * **Prompt Engineering:** For more complex scenarios, you might need to adjust your global or agent-specific prompts to guide DocsGPT on how to interpret and present information from API tool responses. See [Customising Prompts](/Guides/Customising-prompts).
-
-## Using the Configured API Tool in Chat
-
-Once an API action is configured and enabled, DocsGPT's LLM can decide to use it based on your natural language queries.
-
-**Example (based on the phone validation tool in Figure 1):**
-
-1.  **User Query:** "Hey DocsGPT, can you check if +14155555555 is a valid phone number?"
-
-2.  **DocsGPT (LLM Orchestration):**
-    * The LLM analyzes the query.
-    * It matches the intent ("check if ... is a valid phone number") with the description of the "Phone-check" API action.
-    * It identifies `+14155555555` as the value for the `phone` parameter (which was marked as `Filled by LLM` with the description "Phone number to check").
-    * DocsGPT constructs the GET API request.
-3.  **API Tool Execution:**
-    * The API Tool makes the HTTP GET request.
-    * The external API (AbstractAPI) processes the request and returns a JSON response, e.g.:
-        ```json
-        {
-          "phone": "+14155555555",
-          "valid": true,
-          "format": {
-            "international": "+1 415-555-5555",
-            "national": "(415) 555-5555"
-          },
-          "country": {
-            "code": "US",
-            "name": "United States",
-            "prefix": "+1"
-          },
-          "location": "California",
-          "type": "Landline"
-        }
-        ```
-
-4.  **DocsGPT Response Formulation:**
-    * The API Tool passes this JSON response back to the LLM.
-    * The LLM, guided by the tool's description and the user's original query, extracts relevant information and formulates a user-friendly answer.
-    * **DocsGPT Chat Response:** "Yes, +14155555555 appears to be a valid landline phone number in California, United States."
-
-## Advanced Tips and Best Practices
-
-**Clear Description is the Key:** The LLM relies heavily on the `Description` field of the API action and its parameters. Make them unambiguous and action-oriented. Clearly state what the tool does and what kind of input it expects (even if implicitly through parameter descriptions). 
-
-**Iterative Testing:** After configuring an API tool, test it with various phrasings of user queries to ensure the LLM triggers it correctly and interprets the response as expected.
-  
-**Error Handling:**
-    * If an API call fails, the API Tool will return an error message and status code from the `requests` library or the API itself. The LLM may relay this error or try to explain it.
-    * Check DocsGPT's backend logs for more detailed error information if you encounter issues.
-  
-**Security Considerations:**
-    * **API Keys:** Be mindful of API keys and other sensitive credentials. The example image shows an API key directly in the configuration. For production or shared environments avoid exposing configurations with sensitive keys.
-    * **Rate Limits:** Be aware of the rate limits of the APIs you are integrating. Frequent calls from DocsGPT could exceed these limits.
-    * **Data Privacy:** Consider the data privacy implications of sending user query data to third-party APIs.
--   **Idempotency:** For tools that modify data (POST, PUT, DELETE), be aware of whether the API operations are idempotent to avoid unintended consequences from repeated calls if the LLM retries an action.
-
-## Limitations
-
-While powerful, the Generic API Tool has some limitations:
-
--   **Complex Authentication:** Advanced authentication flows like OAuth 2.0 (especially 3-legged OAuth requiring user redirection) or custom signature-based authentication often require custom Python tools.
--   **Multi-Step API Interactions:** If a task requires multiple API calls that depend on each other (e.g., fetch a list, then for each item, fetch details), this kind of complex chaining and logic is better handled by a custom Python tool.
--   **Complex Data Transformations:** If the API response needs significant transformation or processing before being useful to the LLM, a custom Python tool offers more flexibility.
--   **Real-time Streaming (SSE, WebSockets):** The tool is designed for request-response interactions, not for maintaining persistent streaming connections.
-
-For scenarios that exceed these limitations, developing a [Custom Python Tool](/Tools/creating-a-tool) is the recommended approach.

docs/pages/Tools/basics.mdx

@@ -1,92 +0,0 @@
----
-title: Tools Basics - Enhancing DocsGPT Capabilities
-description: Understand what DocsGPT Tools are, how they work, and explore the built-in tools available to extend DocsGPT's functionality.
----
-
-import { Callout } from 'nextra/components';
-import Image from 'next/image';
-import { ToolCards } from '../../components/ToolCards';
-
-# Understanding DocsGPT Tools
-
-DocsGPT Tools are powerful extensions that significantly enhance the capabilities of your DocsGPT application. 
-They allow DocsGPT to move beyond its core function of retrieving information from your documents and enable it to perform actions, 
-interact with external data sources, and integrate with other services. You can find and configure available tools within 
-the "Tools" section of the DocsGPT application settings in the user interface.
-
-## What are Tools?
-
--   **Purpose:** The primary purpose of Tools is to bridge the gap between understanding a user's request (natural language processing by the LLM) and executing a tangible action. This could involve fetching live data from the web, sending notifications, running code snippets, querying databases, or interacting with third-party APIs.
-
--   **LLM as an Orchestrator:** The Large Language Model (LLM) at the heart of DocsGPT is designed to act as an intelligent orchestrator. Based on your query and the declared capabilities of the available tools (defined in their metadata), the LLM decides if a tool is needed, which tool to use, and what parameters to pass to it.
-
--   **Action-Oriented Interactions:** Tools enable more dynamic and action-oriented interactions. For example:
-    * *"What's the latest news on renewable energy?"* - This might trigger a web search tool to fetch current articles.
-    * *"Fetch the order status for customer ID 12345 from our database."* - This could use a database tool.
-    * *"Summarize the content of this webpage and send the summary to the #general channel on Telegram."* - This might involve a web scraping tool followed by a Telegram notification tool.
-
-## Overview of Built-in Tools
-
-DocsGPT includes a suite of pre-built tools designed to expand its capabilities out-of-the-box. Below is an overview of the currently available tools. 
-
-<ToolCards
-  items={[
-    {
-      title: 'API Tool',
-      link: '/Tools/api-tool',
-      description: 'A highly flexible tool that allows DocsGPT to interact with virtually any API without needing to write custom Python code.'
-    },
-    {
-      title: 'Brave Search Tool',
-      link: 'https://github.com/arc53/DocsGPT/blob/main/application/agents/tools/brave.py',
-      description: 'Enables DocsGPT to perform real-time web and image searches using the Brave Search API for up-to-date information.'
-    },
-    {
-      title: 'Cryptoprice Tool',
-      link: 'https://github.com/arc53/DocsGPT/blob/main/application/agents/tools/cryptoprice.py',
-      description: 'Fetches the current price of specified cryptocurrencies.'
-    },
-    {
-      title: 'Ntfy Tool',
-      link: 'https://github.com/arc53/DocsGPT/blob/main/application/agents/tools/ntfy.py',
-      description: 'Allows DocsGPT to send push notifications to Ntfy.sh channels, ideal for alerts and updates.'
-    },
-    {
-      title: 'PostgreSQL Tool',
-      link: 'https://github.com/arc53/DocsGPT/blob/main/application/agents/tools/postgres.py',
-      description: 'Provides capabilities to connect to a PostgreSQL database, execute SQL queries, and retrieve schema information.'
-    },
-    {
-      title: 'Read Webpage Tool', // Renamed from Scraper Tool
-      link: 'https://github.com/arc53/DocsGPT/blob/main/application/agents/tools/read_webpage.py',
-      description: 'Enables DocsGPT to fetch and extract (scrape) textual content from specified web page URLs.'
-    },
-    {
-      title: 'Telegram Tool',
-      link: 'https://github.com/arc53/DocsGPT/blob/main/application/agents/tools/telegram.py',
-      description: 'Allows DocsGPT to send messages or images to Telegram chats via a Telegram Bot.'
-    }
-  ]}
-/>
-
-## Using Tools in DocsGPT (User Perspective)
-
-Interacting with tools in DocsGPT is designed to be intuitive:
-
-1.  **Natural Language Interaction:** As a user, you typically interact with DocsGPT using natural language queries or commands. The LLM within DocsGPT analyzes your input to determine if a specific task can or should be handled by one of the available and configured tools.
-
-2.  **Configuration in UI:**
-    * Tools are generally managed and configured within the DocsGPT application's settings, found under a "Tools" section in the GUI.
-    * For tools that interact with external services (like Brave Search, Telegram, or any service via the API Tool), you might need to provide authentication credentials (e.g., API keys, tokens) or specific endpoint information during the tool's setup in the UI.
-
-3.  **Prompt Engineering for Tools:** While the LLM aims to intelligently use tools, for more complex or reliable agent-like behaviors, you might need to customize the system prompts. Modifying the prompt can guide the LLM on when and how to prioritize or chain tools to achieve specific outcomes, especially if you're building an agent designed to perform a certain sequence of actions every time. For more on this, see [Customising Prompts](/Guides/Customising-prompts).
-
-## Advancing with Tools
-
-Understanding the basics of DocsGPT Tools opens up many possibilities:
-
-* **Leverage the API Tool:** For quick integrations with numerous external services, explore the [API Tool Detailed Guide](/Tools/api-tool).
-* **Develop Custom Tools:** If you have specific needs not covered by built-in tools or the generic API tool, you can develop your own. See our guide on `[Developing Custom Tools](/Tools/creating-a-tool)` (placeholder for now).
-* **Build AI Agents:** Tools are the fundamental building blocks for creating sophisticated AI agents within DocsGPT. Explore how these can be combined by looking into the `[Agents section/tab concept - link to be added once available]`.
-
-By harnessing the power of Tools, you can transform DocsGPT into a more versatile and proactive assistant tailored to your unique workflows.

docs/pages/Tools/creating-a-tool.mdx

@@ -1,186 +0,0 @@
----
-title: 🛠️ Creating a Custom Tool
-description: Learn how to create custom Python tools to extend DocsGPT's functionality and integrate with various services or perform specific actions.
----
-
-import { Callout } from 'nextra/components';
-import { Steps } from 'nextra/components';
-
-# 🛠️ Creating a Custom Python Tool
-
-This guide provides developers with a comprehensive, step-by-step approach to creating their own custom tools for DocsGPT. By developing custom tools, you can significantly extend DocsGPT's capabilities, enabling it to interact with new data sources, services, and perform specialized actions tailored to your unique needs.
-
-## Introduction to Custom Tool Development
-
-### Why Create Custom Tools?
-
-While DocsGPT offers a range of built-in tools and a versatile API Tool, there are many scenarios where a custom Python tool is the best solution:
-
-* **Integrating with Proprietary Systems:** Connect to internal APIs, databases, or services that are not publicly accessible or require complex authentication.
-* **Adding Domain-Specific Functionalities:** Implement logic specific to your industry or use case that isn't covered by general-purpose tools.
-* **Automating Unique Workflows:** Create tools that orchestrate multiple steps or interact with systems in a way unique to your operational needs.
-* **Connecting to Any System with an Accessible Interface:** If you can interact with a system programmatically using Python (e.g., through libraries, SDKs, or direct HTTP requests), you can likely build a DocsGPT tool for it.
-* **Complex Logic or Data Transformation:** When API interactions require intricate logic before sending a request or after receiving a response, or when data needs significant transformation that is difficult for an LLM to handle directly.
-
-### Prerequisites
-
-Before you begin, ensure you have:
-
-* A solid understanding of Python programming.
-* Familiarity with the DocsGPT project structure, particularly the `application/agents/tools/` directory where custom tools reside.
-* Basic knowledge of how APIs work, as many tools involve interacting with external or internal APIs.
-* Your DocsGPT development environment set up. If not, please refer to the [Setting Up a Development Environment](/Deploying/Development-Environment) guide.
-
-## The Anatomy of a DocsGPT Tool
-
-Custom tools in DocsGPT are Python classes that inherit from a base `Tool` class and implement specific methods to define their behavior, capabilities, and configuration needs.
-
-The **foundation** for all custom tools is the abstract base class, located in `application/agents/tools/base.py`. Your custom tool class **must** inherit from this class.
-
-### Essential Methods to Implement
-
-Your custom tool class needs to implement the following methods:
-
-1. **`__init__(self, config: dict)`**
-    
-    - **Purpose:** The constructor for your tool. It's called when DocsGPT initializes the tool.
-    - **Usage:** This method is typically used to receive and store tool-specific configurations passed via the `config` dictionary. This dictionary is populated based on the tool's settings, often configured through the DocsGPT UI or environment variables. For example, you would store API keys, base URLs, or database connection strings here.
-    - **Example** (`brave.py`)**:**
-        ``` python
-        class BraveSearchTool(Tool):
-            def __init__(self, config):
-                self.config = config
-                self.token = config.get("token", "") # API Key for Brave Search
-                self.base_url = "https://api.search.brave.com/res/v1"
-        ```
-
-2. **`execute_action(self, action_name: str, **kwargs) -> dict`**
-    
-    - **Purpose:** This is the workhorse of your tool. The LLM, acting as an agent, calls this method when it decides to use one of the actions your tool provides.
-    - **Parameters:**
-        - `action_name` (str): A string specifying which of the tool's actions to run (e.g., "brave_web_search").
-        - `**kwargs` (dict): A dictionary containing the parameters for that specific action. These parameters are defined in the tool's metadata (`get_actions_metadata()`) and are extracted or inferred by the LLM from the user's query.
-    - **Return Value:** A dictionary containing the result of the action. It's good practice to include keys like:
-        - `status_code` (int): An HTTP-like status code (e.g., 200 for success, 500 for error).
-        - `message` (str): A human-readable message describing the outcome.
-        - `data` (any): The actual data payload returned by the action (if applicable).
-        - `error` (str): An error message if the action failed.
-    - **Example (`read_webpage.py`):**
-        
-        ``` python
-        def execute_action(self, action_name: str, **kwargs) -> str:
-            if action_name != "read_webpage":
-                return f"Error: Unknown action '{action_name}'. This tool only supports 'read_webpage'."
-        
-            url = kwargs.get("url")
-            if not url:
-                return "Error: URL parameter is missing."
-            # ... (logic to fetch and parse webpage) ...
-            try:
-                # ...
-                return markdown_content 
-            except Exception as e:
-                return f"Error processing URL {url}: {e}"
-        ```
-        
-        A more structured return:
-        
-        ``` python
-        # ... inside execute_action
-        try:
-            # ... logic ...
-            return {"status_code": 200, "message": "Webpage read successfully", "data": markdown_content}
-        except Exception as e:
-            return {"status_code": 500, "message": f"Error processing URL {url}", "error": str(e)}
-        ```
-        
-3. **`get_actions_metadata(self) -> list`**
-    
-    - **Purpose:** This method is **critical** for the LLM to understand what your tool can do, when to use it, and what parameters it needs. It effectively advertises your tool's capabilities.
-    - **Return Value:** A list of dictionaries. Each dictionary describes one distinct action the tool can perform and must follow a specific JSON schema structure.
-        - `name` (str): A unique and descriptive name for the action (e.g., `mytool_get_user_details`). It's a common convention to prefix with the tool name to avoid collisions.
-        - `description` (str): A clear, concise, and unambiguous description of what the action does. **Write this for the LLM.** The LLM uses this description to decide if this action is appropriate for a given user query.
-        - `parameters` (dict): A JSON Schema object defining the parameters that the action expects. This schema tells the LLM what arguments are needed, their types, and which are required.
-            - `type`: Should always be `"object"`.
-            - `properties`: A dictionary where each key is a parameter name, and the value is an object defining its `type` (e.g., "string", "integer", "boolean") and `description`.
-            - `required`: A list of strings, where each string is the name of a parameter that is mandatory for the action.
-    - **Example (`postgres.py` - partial):**
-        
-        ``` python
-        def get_actions_metadata(self):
-            return [
-                {
-                    "name": "postgres_execute_sql",
-                    "description": "Execute an SQL query against the PostgreSQL database...",
-                    "parameters": {
-                        "type": "object",
-                        "properties": {
-                            "sql_query": {
-                                "type": "string",
-                                "description": "The SQL query to execute.",
-                            },
-                        },
-                        "required": ["sql_query"],
-                        "additionalProperties": False, # Good practice to prevent unexpected params
-                    },
-                },
-                # ... other actions like postgres_get_schema
-            ]
-        ```
-        
-4. **`get_config_requirements(self) -> dict`**
-    
-    - **Purpose:** Defines the configuration parameters that your tool needs to function (e.g., API keys, specific base URLs, connection strings, default settings). This information can be used by the DocsGPT UI to dynamically render configuration fields for your tool or for validation.
-    - **Return Value:** A dictionary where keys are the configuration item names (which will be keys in the `config` dict passed to `__init__`) and values are dictionaries describing each requirement:
-        - `type` (str): The expected data type of the config value (e.g., "string", "boolean", "integer").
-        - `description` (str): A human-readable description of what this configuration item is for.
-        - `secret` (bool, optional): Set to `True` if the value is sensitive (e.g., an API key) and should be masked or handled specially in UIs. Defaults to `False`.
-    - **Example (`brave.py`):**
-        
-        ``` python
-        def get_config_requirements(self):
-            return {
-                "token": { # This 'token' will be a key in the config dict for __init__
-                    "type": "string",
-                    "description": "Brave Search API key for authentication",
-                    "secret": True
-                },
-            }
-        ```    
-
-## Tool Registration and Discovery
-
-DocsGPT's ToolManager (located in application/agents/tools/tool_manager.py) automatically discovers and loads tools.
-
-As long as your custom tool:
-
-1. Is placed in a Python file within the `application/agents/tools/` directory (and the filename is not `base.py` or starts with `__`).
-2. Correctly inherits from the `Tool` base class.
-3. Implements all the abstract methods (`execute_action`, `get_actions_metadata`, `get_config_requirements`).
-
-The `ToolManager` should be able to load it when DocsGPT starts. 
-
-## Configuration & Secrets Management
-
-- **Configuration Source:** The `config` dictionary passed to your tool's `__init__` method is typically populated from settings defined in the DocsGPT UI (if available for the tool) or from environment variables/configuration files that DocsGPT loads (see [⚙️ App Configuration](/Deploying/DocsGPT-Settings)). The keys in this dictionary should match the names you define in `get_config_requirements()`.
-- **Secrets:** Never hardcode secrets (like API keys or passwords) directly into your tool's Python code. Instead, define them as configuration requirements (using `secret: True` in `get_config_requirements()`) and let DocsGPT's configuration system inject them via the `config` dictionary at runtime. This ensures that secrets are managed securely and are not exposed in your codebase.
-
-## Best Practices for Tool Development
-
-- **Atomicity:** Design tool actions to be as atomic (single, well-defined purpose) as possible. This makes them easier for the LLM to understand and combine.
-- **Clarity in Metadata:** Ensure action names and descriptions in `get_actions_metadata()` are extremely clear, specific, and unambiguous. This is the primary way the LLM understands your tool.
-- **Robust Error Handling:** Implement comprehensive error handling within your `execute_action` logic (and the private methods it calls). Return informative error messages in the result dictionary so the LLM or user can understand what went wrong.
-- **Security:**
-    - Be mindful of the security implications of your tool, especially if it interacts with sensitive systems or can execute arbitrary code/queries.
-    - Validate and sanitize any inputs, especially if they are used to construct database queries or shell commands, to prevent injection attacks.
-- **Performance:** Consider the performance implications of your tool's actions. If an action is slow, it will impact the user experience. Optimize where possible.
-
-## (Optional) Contributing Your Tool
-
-If you develop a custom tool that you believe could be valuable to the broader DocsGPT community and is general-purpose:
-
-1. Ensure it's well-documented (both in code and with clear metadata).
-2. Make sure it adheres to the best practices outlined above.
-3. Consider opening a Pull Request to the [DocsGPT GitHub repository](https://github.com/arc53/DocsGPT) with your new tool, including any necessary documentation updates.
-
-By following this guide, you can create powerful custom tools that extend DocsGPT's capabilities to your specific operational environment.

docs/pages/_app.mdx

@@ -4,7 +4,7 @@ export default function MyApp({ Component, pageProps }) {
   return (
     <>
       <Component {...pageProps} />
-        <DocsGPTWidget showSources={true} apiKey="5d8270cb-735f-484e-9dc9-5b407f24e652" theme="dark" size="medium" />
+        <DocsGPTWidget showSources={true} apiKey="6dd66edf-d374-4904-93af-ab0c6d41ee56" theme="dark" size="medium" />
     </>
   )
 }

docs/pages/_meta.json

@@ -4,8 +4,6 @@
   "quickstart": "Quickstart",
   "Deploying": "Deploying",
   "Models": "Models",
-  "Tools": "Tools",
-  "Agents": "Agents",
   "Extensions": "Extensions",
   "https://gptcloud.arc53.com/": {
     "title": "API",

docs/pages/quickstart.mdx

@@ -73,44 +73,9 @@ The easiest way to launch DocsGPT is using the provided `setup.sh` script. This
 
 ## Launching DocsGPT (Windows)
 
-For Windows users, we provide a PowerShell script that offers the same functionality as the macOS/Linux setup script.
+For Windows users, we recommend following the Docker deployment guide for detailed instructions. Please refer to the [Docker Deployment documentation](/Deploying/Docker-Deploying) for step-by-step instructions on setting up DocsGPT on Windows using Docker.
 
-**Steps:**
-
-1. **Download the DocsGPT Repository:**
-
-   First, you need to download the DocsGPT repository to your local machine. You can do this using Git:
-
-   ```powershell
-   git clone https://github.com/arc53/DocsGPT.git
-   cd DocsGPT
-   ```
-
-2. **Run the `setup.ps1` script:**
-
-   Execute the PowerShell setup script:
-
-   ```powershell
-   PowerShell -ExecutionPolicy Bypass -File .\setup.ps1
-   ```
-
-3. **Follow the interactive setup:**
-
-   Just like the Linux/macOS script, the PowerShell script will guide you through setting DocsGPT.
-   The script will handle environment configuration and start DocsGPT based on your selections.
-
-4. **Access DocsGPT in your browser:**
-
-   Once the setup is complete and Docker containers are running, navigate to [http://localhost:5173/](http://localhost:5173/) in your web browser to access the DocsGPT web application.
-
-5. **Stopping DocsGPT:**
-
-   To stop DocsGPT run the Docker Compose down command displayed at the end of the setup script's execution.
-
-**Important for Windows:** Ensure Docker Desktop is installed and running correctly on your Windows system before proceeding. The script will attempt to start Docker if it's not running, but you may need to start it manually if there are issues.
-
-**Alternative Method:**
-If you prefer a more manual approach, you can follow our [Docker Deployment documentation](/Deploying/Docker-Deploying) for detailed instructions on setting up DocsGPT on Windows using Docker commands directly.
+**Important for Windows:** Ensure Docker Desktop is installed and running correctly on your Windows system before proceeding.
 
 ## Advanced Configuration
 

docs/public/toolIcons/tool_api_tool.svg

@@ -1,6 +0,0 @@
-<?xml version="1.0" encoding="utf-8"?>
-<svg viewBox="1 6 38 28" xmlns="http://www.w3.org/2000/svg">
-  <path d="M3,33.5c-0.827,0-1.5-0.673-1.5-1.5V8c0-0.827,0.673-1.5,1.5-1.5h34c0.827,0,1.5,0.673,1.5,1.5v24 c0,0.827-0.673,1.5-1.5,1.5H3z" style="fill: rgb(7, 106, 255);"/>
-  <path d="M37,7c0.551,0,1,0.449,1,1v24c0,0.551-0.449,1-1,1H3c-0.551,0-1-0.449-1-1V8c0-0.551,0.449-1,1-1 H37 M37,6H3C1.895,6,1,6.895,1,8v24c0,1.105,0.895,2,2,2h34c1.105,0,2-0.895,2-2V8C39,6.895,38.105,6,37,6L37,6z" style="fill: rgb(7, 106, 255);"/>
-  <path d="M 19.296 13.226 C 20.066 13.06 21.108 12.955 22.147 12.955 C 23.772 12.955 25.153 13.185 26.047 14.038 C 26.88 14.766 27.255 15.931 27.255 17.118 C 27.255 18.638 26.798 19.718 26.07 20.489 C 25.196 21.426 23.801 21.842 22.656 21.842 C 22.47 21.842 22.302 21.842 22.115 21.821 L 22.115 27.045 L 19.297 27.045 L 19.297 13.226 L 19.296 13.226 Z M 22.114 19.616 C 22.259 19.637 22.405 19.637 22.571 19.637 C 23.945 19.637 24.55 18.657 24.55 17.347 C 24.55 16.119 24.049 15.162 22.78 15.162 C 22.532 15.162 22.281 15.203 22.114 15.266 L 22.114 19.616 Z M 29.158 12.955 L 31.976 12.955 L 31.976 27.045 L 29.158 27.045 L 29.158 12.955 Z M 15.001 27.045 L 17.887 27.045 L 14.91 12.955 L 11.342 12.955 L 8.024 27.045 L 10.91 27.045 L 11.524 24.227 L 14.408 24.227 L 15.001 27.045 Z M 13 15.547 L 13.068 15.547 C 13.205 16.467 13.409 17.888 13.568 18.745 L 14.021 21.409 L 11.942 21.409 L 12.457 18.746 C 12.614 17.93 12.841 16.488 13 15.547 Z" style="fill: rgb(255, 255, 255);"/>
-</svg>

docs/public/toolIcons/tool_brave.svg

@@ -1 +0,0 @@
-<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" viewBox="0 0 194.18 227.53"><defs><style>.cls-1{fill-rule:evenodd;fill:url(#linear-gradient);}.cls-2{fill:#fff;}</style><linearGradient id="linear-gradient" y1="116.23" x2="194.18" y2="116.23" gradientTransform="matrix(1, 0, 0, -1, 0, 230)" gradientUnits="userSpaceOnUse"><stop offset="0" stop-color="#ff5601"/><stop offset="0.5" stop-color="#ff4000"/><stop offset="1" stop-color="#ff1f01"/></linearGradient></defs><g id="Layer_2" data-name="Layer 2"><g id="Layer_1-2" data-name="Layer 1"><path class="cls-1" d="M187.39,54.58l5.34-13.1s-6.8-7.27-15-15.52S152,22.56,152,22.56L132,0H62.14L42.23,22.56S24.76,17.71,16.51,26s-15,15.52-15,15.52L6.8,54.58,0,74s20,75.65,22.33,84.89c4.61,18.19,7.77,25.22,20.88,34.44S80.1,218.55,84,221s8.74,6.56,13.11,6.56,9.22-4.13,13.11-6.56,27.67-18.43,40.78-27.65,16.26-16.25,20.87-34.44C174.19,149.64,194.18,74,194.18,74Z"/><path class="cls-2" d="M121.85,41c2.91,0,24.51-4.12,24.51-4.12S172,67.8,172,74.41c0,5.47-2.21,7.6-4.8,10.12-.54.53-1.1,1.08-1.66,1.67l-19.2,20.37-.63.64c-1.91,1.92-4.73,4.76-2.74,9.47l.41,1c2.18,5.1,4.87,11.39,1.44,17.78-3.64,6.78-9.89,11.31-13.9,10.56s-13.41-5.66-16.87-7.9S99.6,126.8,99.6,123.35c0-2.89,7.88-7.68,11.71-10,.77-.47,1.37-.83,1.71-1.07l1.88-1.18c3.49-2.17,9.8-6.09,10-7.83.2-2.14.12-2.77-2.69-8.06-.6-1.13-1.3-2.33-2-3.58-2.68-4.61-5.69-9.78-5-13.48.75-4.18,7.3-6.57,12.85-8.6l2-.75,5.78-2.17c5.54-2.07,11.69-4.37,12.71-4.84,1.4-.65,1-1.27-3.22-1.67l-2.06-.21c-5.27-.56-15-1.59-19.71-.28l-3.06.84c-5.31,1.43-11.81,3.19-12.44,4.21-.11.18-.22.33-.32.47-.6.85-1,1.41-.32,5,.19,1.08.6,3.19,1.1,5.81,1.46,7.65,3.75,19.58,4,22.26,0,.38.08.74.13,1.09.36,3,.61,5-2.87,5.77l-.91.21c-3.92.9-9.67,2.22-11.75,2.22s-7.83-1.32-11.76-2.22l-.9-.21c-3.48-.79-3.23-2.78-2.87-5.77,0-.35.09-.71.13-1.09.29-2.68,2.58-14.65,4-22.3.5-2.59.9-4.7,1.1-5.77.66-3.6.27-4.16-.33-5-.1-.14-.21-.29-.32-.47-.62-1-7.13-2.78-12.43-4.21l-3.07-.84C66,58.31,56.25,59.34,51,59.9l-2.06.21c-4.26.4-4.62,1-3.22,1.67,1,.47,7.17,2.77,12.71,4.84l5.78,2.17,2,.75c5.55,2,12.1,4.42,12.85,8.6.67,3.7-2.34,8.87-5,13.48-.72,1.25-1.43,2.45-2,3.58-2.82,5.29-2.9,5.92-2.7,8.06.16,1.74,6.47,5.66,10,7.83.82.5,1.48.92,1.88,1.18s.94.6,1.71,1.06c3.83,2.33,11.71,7.13,11.71,10,0,3.45-11,12.49-14.42,14.73S67.3,145.24,63.29,146,53,142.2,49.39,135.42c-3.43-6.38-.74-12.68,1.44-17.78l.41-1c2-4.71-.83-7.55-2.74-9.47l-.63-.64L28.67,86.2c-.56-.59-1.12-1.14-1.66-1.67-2.59-2.52-4.79-4.65-4.79-10.12,0-6.61,25.6-37.53,25.6-37.53S69.42,41,72.33,41c2.33,0,6.82-1.55,11.49-3.16l3.56-1.21a34.33,34.33,0,0,1,9.71-2,34.33,34.33,0,0,1,9.71,2c1.18.39,2.37.81,3.56,1.21C115,39.45,119.52,41,121.85,41Z"/><path class="cls-2" d="M118.14,150.39c4.57,2.35,7.81,4,9,4.78,1.59,1,.62,2.86-.82,3.88s-20.85,16-22.73,17.69l-.76.68c-1.82,1.64-4.13,3.72-5.77,3.72s-4-2.08-5.77-3.72l-.76-.68c-1.88-1.66-21.28-16.67-22.73-17.69s-2.41-2.89-.82-3.88c1.23-.77,4.47-2.44,9-4.79l4.34-2.24c6.84-3.54,15.37-6.54,16.7-6.54s9.86,3,16.7,6.54Z"/></g></g></svg>

docs/public/toolIcons/tool_cryptoprice.svg

@@ -1 +0,0 @@
-<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 122.88 122.88"><path d="M17.89 0h88.9c8.85 0 16.1 7.24 16.1 16.1v90.68c0 8.85-7.24 16.1-16.1 16.1H16.1c-8.85 0-16.1-7.24-16.1-16.1v-88.9C0 8.05 8.05 0 17.89 0zm57.04 66.96l16.46 4.96c-1.1 4.61-2.84 8.47-5.23 11.56-2.38 3.1-5.32 5.43-8.85 7-3.52 1.57-8.01 2.36-13.45 2.36-6.62 0-12.01-.96-16.21-2.87-4.19-1.92-7.79-5.3-10.83-10.13-3.04-4.82-4.57-11.02-4.57-18.54 0-10.04 2.67-17.76 8.02-23.17 5.36-5.39 12.93-8.09 22.71-8.09 7.65 0 13.68 1.54 18.06 4.64 4.37 3.1 7.64 7.85 9.76 14.27l-16.55 3.66c-.58-1.84-1.19-3.18-1.82-4.03-1.06-1.43-2.35-2.53-3.86-3.3-1.53-.78-3.22-1.16-5.11-1.16-4.27 0-7.54 1.71-9.8 5.12-1.71 2.53-2.57 6.52-2.57 11.94 0 6.73 1.02 11.33 3.07 13.83 2.05 2.49 4.92 3.73 8.63 3.73 3.59 0 6.31-1 8.15-3.03 1.83-1.99 3.16-4.92 3.99-8.75z" fill-rule="evenodd" clip-rule="evenodd"/></svg>

docs/public/toolIcons/tool_postgres.svg

@@ -1,29 +0,0 @@
-<?xml version="1.0" encoding="utf-8"?><!-- Uploaded to: SVG Repo, www.svgrepo.com, Generator: SVG Repo Mixer Tools -->
-<svg width="800px" height="800px" viewBox="-8.78 0 70 70" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:cc="http://creativecommons.org/ns#" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns="http://www.w3.org/2000/svg">
-    <metadata>
-        <rdf:RDF>
-            <cc:Work>
-                <dc:subject>
-                    Data
-                </dc:subject>
-                <dc:identifier>
-                    sql-database-generic
-                </dc:identifier>
-                <dc:title>
-                    SQL Database (Generic)
-                </dc:title>
-                <dc:format>
-                    image/svg+xml
-                </dc:format>
-                <dc:publisher>
-                    Amido Limited
-                </dc:publisher>
-                <dc:creator>
-                    Richard Slater
-                </dc:creator>
-                <dc:type rdf:resource="http://purl.org/dc/dcmitype/StillImage"/>
-            </cc:Work>
-        </rdf:RDF>
-    </metadata>
-    <path d="m 852.97077,1013.9363 c -6.55238,-0.4723 -13.02857,-2.1216 -17.00034,-4.3296 -2.26232,-1.2576 -3.98589,-2.8032 -4.66223,-4.1807 l -0.4024,-0.8196 0,-25.70807 0,-25.7081 0.31843,-0.6465 c 1.42297,-2.889 5.96432,-5.4935 12.30378,-7.0562 2.15195,-0.5305 5.2586,-1.0588 7.79304,-1.3252 2.58797,-0.2721 9.44765,-0.2307 12.02919,0.073 6.86123,0.8061 12.69967,2.6108 16.29768,5.0377 1.38756,0.9359 2.81137,2.4334 3.29371,3.4642 l 0.41358,0.8838 -0.0354,25.6303 -0.0354,25.63047 -0.33195,0.6744 c -0.18257,0.3709 -0.73406,1.1007 -1.22553,1.6216 -2.99181,3.1715 -9.40919,5.5176 -17.8267,6.5172 -1.71567,0.2038 -9.16916,0.3686 -10.92937,0.2417 z m 12.07501,-22.02839 c -0.0252,-0.0657 -1.00472,-0.93831 -2.17671,-1.93922 -1.17199,-1.00091 -2.18138,-1.86687 -2.24309,-1.92436 -0.0617,-0.0575 0.15481,-0.26106 0.48117,-0.45237 0.32635,-0.19131 0.95163,-0.7235 1.3895,-1.18265 1.2805,-1.34272 1.88466,-3.00131 1.88466,-5.17388 0,-2.1388 -0.65162,-3.8645 -1.95671,-5.1818 -1.31533,-1.3278 -2.82554,-1.8983 -5.02486,-1.8983 -3.39007,0 -5.99368,1.9781 -6.82468,5.1851 -0.28586,1.1031 -0.28432,3.33211 0.003,4.31023 0.74941,2.55136 2.79044,4.40434 5.33062,4.83946 0.8596,0.14724 0.97605,0.21071 1.5621,0.85144 0.34829,0.38078 1.06301,1.14085 1.58827,1.68904 l 0.95501,0.9967 2.53878,0 c 1.39633,0 2.51816,-0.0537 2.49296,-0.11939 z m -8.70653,-7.10848 c -0.61119,-0.31868 -0.84225,-0.56599 -1.19079,-1.27453 -0.26919,-0.54724 -0.31522,-0.85851 -0.31824,-2.15197 -0.003,-1.3143 0.0388,-1.5983 0.31987,-2.169 0.45985,-0.9339 1.09355,-1.376 2.07384,-1.4469 1.36454,-0.099 2.15217,0.5707 2.56498,2.1801 0.50612,1.97321 -0.0504,4.07107 -1.26471,4.76729 -0.63707,0.36527 -1.58737,0.40659 -2.18495,0.095 z m -11.25315,3.66269 c 2.66179,-0.5048 4.1728,-2.0528 4.1728,-4.27495 0,-1.97137 -0.97548,-3.12004 -3.6716,-4.32364 -1.54338,-0.689 -2.10241,-1.1215 -2.10241,-1.6268 0,-0.4188 0.53052,-0.8777 1.14813,-0.993 0.60302,-0.1126 2.20237,0.1652 3.14683,0.5467 l 0.79167,0.3198 0,-1.7524 0,-1.7525 -0.85923,-0.1906 c -0.53103,-0.1178 -1.64689,-0.1885 -2.92137,-0.1849 -1.80528,0 -2.15881,0.044 -2.83818,0.3138 -1.98445,0.7878 -2.92613,2.1298 -2.91107,4.1485 0.0141,1.8898 1.01108,3.06864 3.49227,4.12912 1.46399,0.62572 2.05076,1.10218 2.05076,1.66522 0,1.1965 -1.99362,1.34375 -4.10437,0.30315 -0.57805,-0.28498 -1.09739,-0.54137 -1.1541,-0.56976 -0.0567,-0.0284 -0.10311,0.79023 -0.10311,1.81917 0,1.86239 0.002,1.87137 0.33919,1.99974 1.26979,0.48278 4.07626,0.69787 5.52379,0.42335 z m 30.4308,-1.72766 0,-1.58098 -2.40584,0 -2.40583,0 0,-5.43035 0,-5.4303 -2.13089,0 -2.13088,0 0,7.0113 0,7.01131 4.53672,0 4.53672,0 0,-1.58098 z m -14.84745,-27.70503 c 4.23447,-0.2937 7.4086,-0.8482 10.20178,-1.7821 2.78264,-0.9304 4.42643,-2.0562 4.79413,-3.2834 0.14166,-0.4729 0.13146,-0.6523 -0.0665,-1.1708 -0.88775,-2.3245 -5.84694,-4.1104 -13.42493,-4.8345 -3.24154,-0.3098 -9.13671,-0.2094 -12.22745,0.2081 -4.71604,0.6372 -8.54333,1.8208 -10.2451,3.1683 -3.44251,2.726 0.19793,5.7242 8.66397,7.1354 3.67084,0.6119 8.42674,0.828 12.30414,0.559 z" fill="#00bcf2" transform="translate(-830.906 -943.981)"/>
-</svg>

docs/public/toolIcons/tool_read_webpage.svg

@@ -1 +0,0 @@
-<svg xmlns="http://www.w3.org/2000/svg" height="24px" viewBox="0 -960 960 960" width="24px" fill="#e3e3e3"><path d="M480-80q-82 0-155-31.5t-127.5-86Q143-252 111.5-325T80-480q0-83 31.5-155.5t86-127Q252-817 325-848.5T480-880q83 0 155.5 31.5t127 86q54.5 54.5 86 127T880-480q0 82-31.5 155t-86 127.5q-54.5 54.5-127 86T480-80Zm0-82q26-36 45-75t31-83H404q12 44 31 83t45 75Zm-104-16q-18-33-31.5-68.5T322-320H204q29 50 72.5 87t99.5 55Zm208 0q56-18 99.5-55t72.5-87H638q-9 38-22.5 73.5T584-178ZM170-400h136q-3-20-4.5-39.5T300-480q0-21 1.5-40.5T306-560H170q-5 20-7.5 39.5T160-480q0 21 2.5 40.5T170-400Zm216 0h188q3-20 4.5-39.5T580-480q0-21-1.5-40.5T574-560H386q-3 20-4.5 39.5T380-480q0 21 1.5 40.5T386-400Zm268 0h136q5-20 7.5-39.5T800-480q0-21-2.5-40.5T790-560H654q3 20 4.5 39.5T660-480q0 21-1.5 40.5T654-400Zm-16-240h118q-29-50-72.5-87T584-782q18 33 31.5 68.5T638-640Zm-234 0h152q-12-44-31-83t-45-75q-26 36-45 75t-31 83Zm-200 0h118q9-38 22.5-73.5T376-782q-56 18-99.5 55T204-640Z"/></svg>

docs/public/toolIcons/tool_telegram.svg

@@ -1,10 +0,0 @@
-<svg width="24" height="25" viewBox="0 0 24 25" fill="none" xmlns="http://www.w3.org/2000/svg">
-<path d="M12 0.5C8.81812 0.5 5.76375 1.76506 3.51562 4.01469C1.2652 6.26522 0.000643966 9.31734 0 12.5C0 15.6813 1.26562 18.7357 3.51562 20.9853C5.76375 23.2349 8.81812 24.5 12 24.5C15.1819 24.5 18.2362 23.2349 20.4844 20.9853C22.7344 18.7357 24 15.6813 24 12.5C24 9.31869 22.7344 6.26431 20.4844 4.01469C18.2362 1.76506 15.1819 0.5 12 0.5Z" fill="url(#paint0_linear_5586_9958)"/>
-<path d="M5.43282 12.373C8.93157 10.849 11.2641 9.8443 12.4303 9.3588C15.7641 7.97261 16.4559 7.73186 16.9078 7.7237C17.0072 7.72211 17.2284 7.74667 17.3728 7.86339C17.4928 7.96183 17.5266 8.09495 17.5434 8.18842C17.5584 8.2818 17.5791 8.49461 17.5622 8.66074C17.3822 10.5582 16.6003 15.1629 16.2028 17.2882C16.0359 18.1874 15.7041 18.4889 15.3834 18.5184C14.6859 18.5825 14.1572 18.0579 13.4822 17.6155C12.4266 16.9231 11.8303 16.4922 10.8047 15.8167C9.6197 15.0359 10.3884 14.6067 11.0634 13.9055C11.2397 13.7219 14.3109 10.9291 14.3691 10.6758C14.3766 10.6441 14.3841 10.526 14.3128 10.4637C14.2434 10.4013 14.1403 10.4227 14.0653 10.4395C13.9584 10.4635 12.2728 11.5788 9.00282 13.7851C8.52469 14.114 8.09157 14.2743 7.70157 14.2659C7.27407 14.2567 6.44907 14.0236 5.83595 13.8245C5.08595 13.5802 4.48782 13.451 4.54032 13.036C4.56657 12.82 4.8647 12.599 5.43282 12.373Z" fill="white"/>
-<defs>
-<linearGradient id="paint0_linear_5586_9958" x1="1200" y1="0.5" x2="1200" y2="2400.5" gradientUnits="userSpaceOnUse">
-<stop stop-color="#2AABEE"/>
-<stop offset="1" stop-color="#229ED9"/>
-</linearGradient>
-</defs>
-</svg>

extensions/react-widget/package.json

@@ -1,6 +1,6 @@
 {
   "name": "docsgpt",
-  "version": "0.5.1",
+  "version": "0.5.0",
   "private": false,
   "description": "DocsGPT 🦖 is an innovative open-source tool designed to simplify the retrieval of information from project documentation using advanced GPT models 🤖.",
   "source": "./src/index.html",

extensions/react-widget/src/components/SearchBar.tsx

@@ -4,19 +4,19 @@ import { WidgetCore } from './DocsGPTWidget';
 import { SearchBarProps } from '@/types';
 import { getSearchResults } from '../requests/searchAPI';
 import { Result } from '@/types';
+import MarkdownIt from 'markdown-it';
 import { getOS, processMarkdownString } from '../utils/helper';
 import DOMPurify from 'dompurify';
-import {
-    CodeIcon,
+import { 
+    CodeIcon, 
     TextAlignLeftIcon,
     HeadingIcon,
-    ReaderIcon,
-    ListBulletIcon,
-    QuoteIcon
+    ReaderIcon, 
+    ListBulletIcon, 
+    QuoteIcon 
 } from '@radix-ui/react-icons';
 const themes = {
     dark: {
-        name: 'dark',
         bg: '#202124',
         text: '#EDEDED',
         primary: {
@@ -29,7 +29,6 @@ const themes = {
         }
     },
     light: {
-        name: 'light',
         bg: '#EAEAEA',
         text: '#171717',
         primary: {
@@ -45,16 +44,15 @@ const themes = {
 
 const GlobalStyle = createGlobalStyle`
   .highlight {
-    color: ${props => props.theme.name === 'dark' ? '#4B9EFF' : '#0066CC'};
-    font-weight: 500;
+    color:#007EE6;
   }
 `;
 
 const loadGeistFont = () => {
-    const link = document.createElement('link');
-    link.href = 'https://fonts.googleapis.com/css2?family=Geist:wght@100..900&display=swap';
-    link.rel = 'stylesheet';
-    document.head.appendChild(link);
+  const link = document.createElement('link');
+  link.href = 'https://fonts.googleapis.com/css2?family=Geist:wght@100..900&display=swap'; 
+  link.rel = 'stylesheet';
+  document.head.appendChild(link);
 };
 
 const Main = styled.div`
@@ -83,27 +81,12 @@ const Container = styled.div`
     position: relative;
     display: inline-block;
 `
-const SearchOverlay = styled.div`
-    position: fixed;
-    top: 0;
-    left: 0;
-    width: 100%;
-    height: 100%;
-    background-color: #0000001A;
-    backdrop-filter: blur(8px);
-    -webkit-backdrop-filter: blur(8px);
-    z-index: 99;
-`;
-
-
 const SearchResults = styled.div`
     position: fixed;
     display: flex;
     flex-direction: column;
-    background-color: ${props => props.theme.name === 'dark' ? 
-        'rgba(0, 0, 0, 0.15)' : 
-        'rgba(255, 255, 255, 0.4)'};
-    border: 1px solid rgba(255, 255, 255, 0.18);
+    background-color: ${props => props.theme.primary.bg};
+    border: 1px solid ${props => props.theme.bg};
     border-radius: 15px;
     padding: 8px 0px 8px 0px;
     width: 792px;
@@ -114,12 +97,8 @@ const SearchResults = styled.div`
     top: 50%;
     transform: translate(-50%, -50%);
     color: ${props => props.theme.primary.text};
-    
-    box-shadow: 0 8px 32px 0 rgba(31, 38, 135, 0.37);
-    backdrop-filter: blur(82px);
-    -webkit-backdrop-filter: blur(82px);
-    border-radius: 10px;
-    
+    box-shadow: 0 4px 6px rgba(0, 0, 0, 0.1), 0 2px 4px rgba(0, 0, 0, 0.1);
+    backdrop-filter: blur(16px);
     box-sizing: border-box;
 
     @media only screen and (max-width: 768px) {
@@ -163,33 +142,6 @@ const ContentWrapper = styled.div`
     flex-direction: column;
     gap: 12px; 
 `;
-
-
-
-const ResultWrapper = styled.div`
-    display: flex;
-    align-items: flex-start;
-    width: 100%;
-    box-sizing: border-box;
-    padding: 8px 16px;
-    cursor: pointer;
-    background-color: transparent;
-    font-family: 'Geist', sans-serif;
-    border-radius: 8px;
-
-    word-wrap: break-word;
-    overflow-wrap: break-word;
-    word-break: break-word;
-    white-space: normal;
-    overflow: hidden;
-    text-overflow: ellipsis;
-
-    &:hover {
-        backdrop-filter: blur(8px);
-        -webkit-backdrop-filter: blur(8px);
-    }
-`;
-
 const Content = styled.div`
     display: flex;
     margin-left: 8px;
@@ -199,10 +151,9 @@ const Content = styled.div`
     font-size: 15px;
     color: ${props => props.theme.primary.text};
     line-height: 1.6;
-    border-left: 2px solid ${props => props.theme.primary.text}CC;
+    border-left: 2px solid #585858;
     overflow: hidden;
-    
-`;
+`
 const ContentSegment = styled.div`
     display: flex;
     align-items: flex-start;
@@ -214,6 +165,80 @@ const ContentSegment = styled.div`
     text-overflow: ellipsis;
 `
 
+const ResultWrapper = styled.div`
+    display: flex;
+    align-items: flex-start;
+    width: 100%;
+    box-sizing: border-box;
+    padding: 8px 16px;
+    cursor: pointer;
+    background-color: ${props => props.theme.primary.bg};
+    font-family: 'Geist', sans-serif;
+    transition: background-color 0.2s;
+    border-radius: 8px;
+
+    word-wrap: break-word;
+    overflow-wrap: break-word;
+    word-break: break-word;
+    white-space: normal;
+    overflow: hidden;
+    text-overflow: ellipsis;
+
+    &:hover {
+        background-color: ${props => props.theme.bg};
+    }
+`
+const Markdown = styled.div`
+line-height:18px;
+font-size: 11px;
+white-space: pre-wrap;
+ pre {
+      padding: 8px;
+      width: 90%;
+      font-size: 11px;
+      border-radius: 6px;
+      overflow-x: auto;
+      background-color: #1B1C1F;
+      color: #fff ;
+    }
+
+    h1,h2 {
+      font-size: 14px;
+      font-weight: 600;
+      color: ${(props) => props.theme.text};
+      opacity: 0.8;
+    }
+
+
+    h3 {
+      font-size: 12px;
+    }
+
+    p {
+      margin: 0px;
+      line-height: 1.35rem;
+      font-size: 11px;
+    }
+
+    code:not(pre code) {
+      border-radius: 6px;
+      padding: 2px 2px;
+      margin: 2px;
+      font-size: 9px;
+      display: inline;
+      background-color: #646464;
+      color: #fff ;
+    }
+    img{
+        max-width: 50%;
+    }
+    code {
+      overflow-x: auto;
+    }
+    a{
+        color: #007ee6;
+    }
+`
 const Toolkit = styled.kbd`
     position: absolute;
     right: 4px;
@@ -234,8 +259,8 @@ const Toolkit = styled.kbd`
 `
 const Loader = styled.div`
   margin: 2rem auto;
-  border: 4px solid ${props => props.theme.name === 'dark' ? 'rgba(255, 255, 255, 0.2)' : 'rgba(0, 0, 0, 0.1)'};
-  border-top: 4px solid ${props => props.theme.name === 'dark' ? '#FFFFFF' : props.theme.primary.bg};
+  border: 4px solid ${props => props.theme.secondary.text};
+  border-top: 4px solid ${props => props.theme.primary.bg};
   border-radius: 50%;
   width: 12px;
   height: 12px;
@@ -255,8 +280,7 @@ const NoResults = styled.div`
   margin-top: 2rem;
   text-align: center;
   font-size: 14px;
-  color: ${props => props.theme.name === 'dark' ? '#E0E0E0' : '#505050'};
-  font-weight: 500;
+  color: #888;
 `;
 const AskAIButton = styled.button`
     display: flex;
@@ -269,35 +293,25 @@ const AskAIButton = styled.button`
     height: 50px;
     padding: 8px 24px;
     border: none;
-    border-radius: 8px;
+    border-radius: 6px;
+    background-color: ${props => props.theme.bg};
     color: ${props => props.theme.text}; 
     cursor: pointer;
+    transition: background-color 0.2s, box-shadow 0.2s;
     font-size: 16px;
-    backdrop-filter: blur(16px);
-    -webkit-backdrop-filter: blur(16px);
-    background-color: ${props => props.theme.name === 'dark' ? 
-        'rgba(255, 255, 255, 0.05)' : 
-        'rgba(0, 0, 0, 0.03)'};
 
     &:hover {
-        backdrop-filter: blur(20px);
-        -webkit-backdrop-filter: blur(20px);
-        background-color: ${props => props.theme.name === 'dark' ? 
-            'rgba(255, 255, 255, 0.1)' : 
-            'rgba(0, 0, 0, 0.06)'}; 
+        opacity: 0.8;
     }
-`;
-
+`
 const SearchHeader = styled.div`
     display: flex;
     align-items: center;
     gap: 8px;
     margin-bottom: 12px;
     padding-bottom: 12px;
-    border-bottom: 1px solid ${props => props.theme.name === 'dark' ? '#FFFFFF24' : 'rgba(0, 0, 0, 0.14)'};
-`;
-
-
+    border-bottom: 1px solid ${props => props.theme.bg};
+`
 
 const TextField = styled.input`
     width: calc(100% - 32px);
@@ -313,16 +327,8 @@ const TextField = styled.input`
     &:focus {
         border-color: none;
     }
-
-    &::placeholder {
-        color: ${props => props.theme.name === 'dark' ? 'rgba(255, 255, 255, 0.6)' : 'rgba(0, 0, 0, 0.5)'} !important;
-        opacity: 100%; /* Force opacity to ensure placeholder is visible */
-        font-weight: 500;
-    }
 `
 
-
-
 const EscapeInstruction = styled.kbd`
     display: flex;
     align-items: center;
@@ -331,21 +337,17 @@ const EscapeInstruction = styled.kbd`
     padding: 4px 8px;
     border-radius: 4px;
     background-color: transparent;
-    border: 1px solid ${props => props.theme.name === 'dark' ? 
-        'rgba(237, 237, 237, 0.6)' : 
-        'rgba(23, 23, 23, 0.6)'};
-    color: ${props => props.theme.name === 'dark' ? '#EDEDED' : '#171717'};
+    border: 1px solid ${props => props.theme.secondary.text};
+    color: ${props => props.theme.text};
     font-size: 12px;
     font-family: 'Geist', sans-serif;
     white-space: nowrap;
     cursor: pointer;
     width: fit-content;
-    -webkit-appearance: none;
-    -moz-appearance: none;
-    appearance: none;
-`;
-
-
+    &:hover {
+        background-color: rgba(255, 255, 255, 0.1);
+    }
+`
 export const SearchBar = ({
     apiKey = "74039c6d-bff7-44ce-ae55-2973cbf13837",
     apiHost = "https://gptcloud.arc53.com",
@@ -365,7 +367,7 @@ export const SearchBar = ({
     const abortControllerRef = React.useRef<AbortController | null>(null);
     const browserOS = getOS();
     const isTouch = 'ontouchstart' in window;
-
+    
     const getKeyboardInstruction = () => {
         if (isResultVisible) return "Enter";
         return browserOS === 'mac' ? '⌘ + K' : 'Ctrl + K';
@@ -392,7 +394,7 @@ export const SearchBar = ({
             }
         };
 
-
+     
         document.addEventListener('mousedown', handleClickOutside);
         document.addEventListener('keydown', handleKeyDown);
         return () => {
@@ -402,34 +404,33 @@ export const SearchBar = ({
     }, []);
 
     React.useEffect(() => {
-    if (!input) {
-        setResults([]);
-        setLoading(false);
-        return;
-    }
-    setLoading(true);
-    if (debounceTimeout.current) {
-        clearTimeout(debounceTimeout.current);
-    }
+        if (!input) {
+            setResults([]);
+            return;
+        }
+        setLoading(true);
+        if (debounceTimeout.current) {
+            clearTimeout(debounceTimeout.current);
+        }
 
-    if (abortControllerRef.current) {
-        abortControllerRef.current.abort();
-    }
-    const abortController = new AbortController();
-    abortControllerRef.current = abortController;
+        if (abortControllerRef.current) {
+            abortControllerRef.current.abort();
+        }
+        const abortController = new AbortController();
+        abortControllerRef.current = abortController;
 
-    debounceTimeout.current = setTimeout(() => {
-        getSearchResults(input, apiKey, apiHost, abortController.signal)
-            .then((data) => setResults(data))
-            .catch((err) => !abortController.signal.aborted && console.log(err))
-            .finally(() => setLoading(false));
-    }, 500);
+        debounceTimeout.current = setTimeout(() => {
+            getSearchResults(input, apiKey, apiHost, abortController.signal)
+                .then((data) => setResults(data))
+                .catch((err) => !abortController.signal.aborted && console.log(err))
+                .finally(() => setLoading(false));
+        }, 500);
 
-    return () => {
-        abortController.abort();
-        clearTimeout(debounceTimeout.current ?? undefined);
-    };
-}, [input])
+        return () => {
+            abortController.abort();
+            clearTimeout(debounceTimeout.current ?? undefined);
+        };
+    }, [input])
 
     const handleKeyDown = (event: React.KeyboardEvent<HTMLInputElement>) => {
         if (event.key === 'Enter') {
@@ -461,8 +462,6 @@ export const SearchBar = ({
                     </SearchButton>
                     {
                         isResultVisible && (
-                            <>
-                            <SearchOverlay onClick={() => setIsResultVisible(false)} />
                             <SearchResults>
                                 <SearchHeader>
                                     <TextField
@@ -478,8 +477,8 @@ export const SearchBar = ({
                                     </EscapeInstruction>
                                 </SearchHeader>
                                 <AskAIButton onClick={openWidget}>
-                                    <img
-                                        src="https://d3dg1063dc54p9.cloudfront.net/cute-docsgpt.png"
+                                    <img 
+                                        src="https://d3dg1063dc54p9.cloudfront.net/cute-docsgpt.png" 
                                         alt="DocsGPT"
                                         width={24}
                                         height={24}
@@ -540,7 +539,6 @@ export const SearchBar = ({
                                     )}
                                 </SearchResultsScroll>
                             </SearchResults>
-                            </>
                         )
                     }
                     {

frontend/.env.development

@@ -1,4 +1,3 @@
 # Please put appropriate value
-VITE_BASE_URL=http://localhost:5173
-VITE_API_HOST=http://127.0.0.1:7091
+VITE_API_HOST=http://0.0.0.0:7091
 VITE_API_STREAMING=true

frontend/package.json

@@ -19,56 +19,52 @@
     ]
   },
   "dependencies": {
-    "@reduxjs/toolkit": "^2.8.2",
+    "@reduxjs/toolkit": "^2.5.1",
     "chart.js": "^4.4.4",
-    "clsx": "^2.1.1",
-    "copy-to-clipboard": "^3.3.3",
     "i18next": "^24.2.0",
     "i18next-browser-languagedetector": "^8.0.2",
-    "lodash": "^4.17.21",
-    "mermaid": "^11.6.0",
     "prop-types": "^15.8.1",
-    "react": "^19.1.0",
+    "react": "^18.2.0",
     "react-chartjs-2": "^5.3.0",
-    "react-dom": "^19.0.0",
-    "react-dropzone": "^14.3.8",
+    "react-copy-to-clipboard": "^5.1.0",
+    "react-dom": "^18.3.1",
+    "react-helmet": "^6.1.0",
+    "react-dropzone": "^14.3.5",
     "react-i18next": "^15.4.0",
     "react-markdown": "^9.0.1",
-    "react-redux": "^9.2.0",
-    "react-router-dom": "^7.6.1",
-    "react-syntax-highlighter": "^15.6.1",
+    "react-redux": "^8.0.5",
+    "react-router-dom": "^7.1.1",
+    "react-syntax-highlighter": "^15.5.0",
     "rehype-katex": "^7.0.1",
     "remark-gfm": "^4.0.0",
-    "remark-math": "^6.0.0",
-    "tailwind-merge": "^3.3.1"
+    "remark-math": "^6.0.0"
   },
   "devDependencies": {
-    "@tailwindcss/postcss": "^4.1.10",
-    "@types/lodash": "^4.17.20",
-    "@types/mermaid": "^9.1.0",
-    "@types/react": "^19.1.8",
-    "@types/react-dom": "^19.0.0",
+    "@types/react": "^18.0.27",
+    "@types/react-dom": "^18.3.0",
+    "@types/react-helmet": "^6.1.11",
     "@types/react-syntax-highlighter": "^15.5.13",
     "@typescript-eslint/eslint-plugin": "^5.51.0",
     "@typescript-eslint/parser": "^5.62.0",
     "@vitejs/plugin-react": "^4.3.4",
+    "autoprefixer": "^10.4.13",
     "eslint": "^8.57.1",
-    "eslint-config-prettier": "^10.1.5",
+    "eslint-config-prettier": "^9.1.0",
     "eslint-config-standard-with-typescript": "^34.0.0",
     "eslint-plugin-import": "^2.31.0",
     "eslint-plugin-n": "^15.7.0",
     "eslint-plugin-prettier": "^5.2.1",
     "eslint-plugin-promise": "^6.6.0",
-    "eslint-plugin-react": "^7.37.5",
+    "eslint-plugin-react": "^7.37.3",
     "eslint-plugin-unused-imports": "^4.1.4",
     "husky": "^8.0.0",
     "lint-staged": "^15.3.0",
     "postcss": "^8.4.49",
-    "prettier": "^3.5.3",
-    "prettier-plugin-tailwindcss": "^0.6.13",
-    "tailwindcss": "^4.1.10",
-    "typescript": "^5.8.3",
-    "vite": "^6.3.5",
-    "vite-plugin-svgr": "^4.3.0"
+    "prettier": "^3.4.2",
+    "prettier-plugin-tailwindcss": "^0.6.9",
+    "tailwindcss": "^3.4.17",
+    "typescript": "^5.7.2",
+    "vite": "^5.4.14",
+    "vite-plugin-svgr": "^4.2.0"
   }
 }

frontend/postcss.config.cjs

@@ -1,5 +1,6 @@
 module.exports = {
   plugins: {
-    '@tailwindcss/postcss': {},
+    tailwindcss: {},
+    autoprefixer: {},
   },
 }