feat: skip empty fields in mcp tool call + improve error handling and response

application/agents/base.py

@@ -142,28 +142,28 @@ class BaseAgent(ABC):
         tool_id, action_name, call_args = parser.parse_args(call)
 
         call_id = getattr(call, "id", None) or str(uuid.uuid4())
-        
+
         # Check if parsing failed
         if tool_id is None or action_name is None:
             error_message = f"Error: Failed to parse LLM tool call. Tool name: {getattr(call, 'name', 'unknown')}"
             logger.error(error_message)
-            
+
             tool_call_data = {
                 "tool_name": "unknown",
                 "call_id": call_id,
-                "action_name": getattr(call, 'name', 'unknown'),
+                "action_name": getattr(call, "name", "unknown"),
                 "arguments": call_args or {},
                 "result": f"Failed to parse tool call. Invalid tool name format: {getattr(call, 'name', 'unknown')}",
             }
             yield {"type": "tool_call", "data": {**tool_call_data, "status": "error"}}
             self.tool_calls.append(tool_call_data)
             return f"Failed to parse tool call.", call_id
-        
+
         # Check if tool_id exists in available tools
         if tool_id not in tools_dict:
             error_message = f"Error: Tool ID '{tool_id}' extracted from LLM call not found in available tools_dict. Available IDs: {list(tools_dict.keys())}"
             logger.error(error_message)
-            
+
             # Return error result
             tool_call_data = {
                 "tool_name": "unknown",
@@ -175,7 +175,7 @@ class BaseAgent(ABC):
             yield {"type": "tool_call", "data": {**tool_call_data, "status": "error"}}
             self.tool_calls.append(tool_call_data)
             return f"Tool with ID {tool_id} not found.", call_id
-        
+
         tool_call_data = {
             "tool_name": tools_dict[tool_id]["name"],
             "call_id": call_id,

application/agents/tools/mcp_tool.py

@@ -283,7 +283,14 @@ class MCPTool(Tool):
         """
         self._ensure_valid_session()
 
-        call_params = {"name": action_name, "arguments": kwargs}
+        # Skipping empty/None values - letting the server use defaults
+
+        cleaned_kwargs = {}
+        for key, value in kwargs.items():
+            if value == "" or value is None:
+                continue
+            cleaned_kwargs[key] = value
+        call_params = {"name": action_name, "arguments": cleaned_kwargs}
         try:
             result = self._make_mcp_request("tools/call", call_params)
             return result

application/core/settings.py

@@ -26,7 +26,7 @@ class Settings(BaseSettings):
         "gpt-4o-mini": 128000,
         "gpt-3.5-turbo": 4096,
         "claude-2": 1e5,
-        "gemini-2.0-flash-exp": 1e6,
+        "gemini-2.5-flash": 1e6,
     }
     UPLOAD_FOLDER: str = "inputs"
     PARSE_PDF_AS_IMAGE: bool = False

application/llm/google_ai.py

@@ -151,15 +151,8 @@ class GoogleLLM(BaseLLM):
 
             if role == "assistant":
                 role = "model"
-            elif role == "system":
-                continue
             elif role == "tool":
-                continue
-            elif role not in ["user", "model"]:
-                logging.warning(
-                    f"GoogleLLM: Converting unsupported role '{role}' to 'user'"
-                )
-                role = "user"
+                role = "model"
 
             parts = []
             if role and content is not None:

application/llm/handlers/base.py

@@ -205,7 +205,6 @@ class LLMHandler(ABC):
                     except StopIteration as e:
                         tool_response, call_id = e.value
                         break
-
                 updated_messages.append(
                     {
                         "role": "assistant",
@@ -222,17 +221,36 @@ class LLMHandler(ABC):
                 )
 
                 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,
-                    }
+                error_call = ToolCall(
+                    id=call.id, name=call.name, arguments=call.arguments
                 )
+                error_response = f"Error executing tool: {str(e)}"
+                error_message = self.create_tool_message(error_call, error_response)
+                updated_messages.append(error_message)
 
+                call_parts = call.name.split("_")
+                if len(call_parts) >= 2:
+                    tool_id = call_parts[-1]  # Last part is tool ID (e.g., "1")
+                    action_name = "_".join(call_parts[:-1])
+                    tool_name = tools_dict.get(tool_id, {}).get("name", "unknown_tool")
+                    full_action_name = f"{action_name}_{tool_id}"
+                else:
+                    tool_name = "unknown_tool"
+                    action_name = call.name
+                    full_action_name = call.name
+                yield {
+                    "type": "tool_call",
+                    "data": {
+                        "tool_name": tool_name,
+                        "call_id": call.id,
+                        "action_name": full_action_name,
+                        "arguments": call.arguments,
+                        "error": error_response,
+                        "status": "error",
+                    },
+                }
         return updated_messages
 
     def handle_non_streaming(
@@ -263,13 +281,11 @@ class LLMHandler(ABC):
                 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(

application/llm/handlers/google.py

@@ -17,7 +17,6 @@ class GoogleLLMHandler(LLMHandler):
                 finish_reason="stop",
                 raw_response=response,
             )
-
         if hasattr(response, "candidates"):
             parts = response.candidates[0].content.parts if response.candidates else []
             tool_calls = [
@@ -41,7 +40,6 @@ class GoogleLLMHandler(LLMHandler):
                 finish_reason="tool_calls" if tool_calls else "stop",
                 raw_response=response,
             )
-
         else:
             tool_calls = []
             if hasattr(response, "function_call"):
@@ -61,14 +59,16 @@ class GoogleLLMHandler(LLMHandler):
 
     def create_tool_message(self, tool_call: ToolCall, result: Any) -> Dict:
         """Create Google-style tool message."""
-        from google.genai import types
 
         return {
-            "role": "tool",
+            "role": "model",
             "content": [
-                types.Part.from_function_response(
-                    name=tool_call.name, response={"result": result}
-                ).to_json_dict()
+                {
+                    "function_response": {
+                        "name": tool_call.name,
+                        "response": {"result": result},
+                    }
+                }
             ],
         }
 

frontend/src/conversation/ConversationBubble.tsx

@@ -1,6 +1,6 @@
 import 'katex/dist/katex.min.css';
 
-import { forwardRef, Fragment, useRef, useState, useEffect } from 'react';
+import { forwardRef, Fragment, useEffect, useRef, useState } from 'react';
 import { useTranslation } from 'react-i18next';
 import ReactMarkdown from 'react-markdown';
 import { useSelector } from 'react-redux';
@@ -12,12 +12,13 @@ import {
 import rehypeKatex from 'rehype-katex';
 import remarkGfm from 'remark-gfm';
 import remarkMath from 'remark-math';
-import DocumentationDark from '../assets/documentation-dark.svg';
+
 import ChevronDown from '../assets/chevron-down.svg';
 import Cloud from '../assets/cloud.svg';
 import DocsGPT3 from '../assets/cute_docsgpt3.svg';
 import Dislike from '../assets/dislike.svg?react';
 import Document from '../assets/document.svg';
+import DocumentationDark from '../assets/documentation-dark.svg';
 import Edit from '../assets/edit.svg';
 import Like from '../assets/like.svg?react';
 import Link from '../assets/link.svg';
@@ -761,7 +762,11 @@ function ToolCalls({ toolCalls }: { toolCalls: ToolCallsType[] }) {
                         Response
                       </span>{' '}
                       <CopyButton
-                        textToCopy={JSON.stringify(toolCall.result, null, 2)}
+                        textToCopy={
+                          toolCall.status === 'error'
+                            ? toolCall.error || 'Unknown error'
+                            : JSON.stringify(toolCall.result, null, 2)
+                        }
                       />
                     </p>
                     {toolCall.status === 'pending' && (
@@ -779,6 +784,16 @@ function ToolCalls({ toolCalls }: { toolCalls: ToolCallsType[] }) {
                         </span>
                       </p>
                     )}
+                    {toolCall.status === 'error' && (
+                      <p className="dark:bg-raisin-black rounded-b-2xl p-2 font-mono text-sm break-words">
+                        <span
+                          className="leading-[23px] text-red-500 dark:text-red-400"
+                          style={{ fontFamily: 'IBMPlexMono-Medium' }}
+                        >
+                          {toolCall.error}
+                        </span>
+                      </p>
+                    )}
                   </div>
                 </div>
               </Accordion>

frontend/src/conversation/types/index.ts

@@ -4,5 +4,6 @@ export type ToolCallsType = {
   call_id: string;
   arguments: Record<string, any>;
   result?: Record<string, any>;
-  status?: 'pending' | 'completed';
+  error?: string;
+  status?: 'pending' | 'completed' | 'error';
 };