# DocsGPT > DocsGPT is an open-source platform for building AI agents and assistants with document retrieval, tools, and multi-model support. This file is a curated map of DocsGPT documentation for LLM and agent use. Prioritize Core, Deploying, and Agents for implementation tasks. ## Core - [Docs Home](https://docs.docsgpt.cloud/): Main documentation landing page. - [Quickstart](https://docs.docsgpt.cloud/quickstart): Fastest path to run DocsGPT locally. - [Architecture](https://docs.docsgpt.cloud/Guides/Architecture): High-level system architecture. - [Development Environment](https://docs.docsgpt.cloud/Deploying/Development-Environment): Backend and frontend local setup. - [DocsGPT Settings](https://docs.docsgpt.cloud/Deploying/DocsGPT-Settings): Environment variables and core app configuration. ## Deploying - [Docker Deployment](https://docs.docsgpt.cloud/Deploying/Docker-Deploying): Run DocsGPT with Docker and Docker Compose. - [Kubernetes Deployment](https://docs.docsgpt.cloud/Deploying/Kubernetes-Deploying): Deploy DocsGPT on Kubernetes clusters. - [Hosting DocsGPT](https://docs.docsgpt.cloud/Deploying/Hosting-the-app): Hosting overview with cloud options. ## Agents - [Agent Basics](https://docs.docsgpt.cloud/Agents/basics): Core concepts for building and managing agents. - [Workflow Nodes](https://docs.docsgpt.cloud/Agents/nodes): Node types and behavior in agent workflows. - [Agent API](https://docs.docsgpt.cloud/Agents/api): Programmatic agent interaction (streaming and non-streaming). - [Agent Webhooks](https://docs.docsgpt.cloud/Agents/webhooks): Trigger and automate agents with webhooks. ## Tools - [Tools Basics](https://docs.docsgpt.cloud/Tools/basics): How tools extend agent capabilities. - [Generic API Tool](https://docs.docsgpt.cloud/Tools/api-tool): Configure API calls without custom code. - [Creating a Custom Tool](https://docs.docsgpt.cloud/Tools/creating-a-tool): Build custom Python tools for DocsGPT. ## Models - [Cloud LLM Providers](https://docs.docsgpt.cloud/Models/cloud-providers): Configure hosted model providers. - [Local Inference](https://docs.docsgpt.cloud/Models/local-inference): Connect DocsGPT to local inference backends. - [Embeddings](https://docs.docsgpt.cloud/Models/embeddings): Select and configure embedding models. ## Extensions - [API Keys for Integrations](https://docs.docsgpt.cloud/Extensions/api-key-guide): Generate and use DocsGPT API keys. - [Chat Widget](https://docs.docsgpt.cloud/Extensions/chat-widget): Embed the DocsGPT chat widget. - [Search Widget](https://docs.docsgpt.cloud/Extensions/search-widget): Embed the DocsGPT search widget. - [Chrome Extension](https://docs.docsgpt.cloud/Extensions/Chrome-extension): Install and use the browser extension. - [Chatwoot Extension](https://docs.docsgpt.cloud/Extensions/Chatwoot-extension): Integrate DocsGPT with Chatwoot. ## Integrations - [Google Drive Connector](https://docs.docsgpt.cloud/Guides/Integrations/google-drive-connector): Ingest and sync files from Google Drive. ## Optional - [Customizing Prompts](https://docs.docsgpt.cloud/Guides/Customising-prompts): Template-based prompt customization. - [How to Train on Other Documentation](https://docs.docsgpt.cloud/Guides/How-to-train-on-other-documentation): Add additional documentation sources. - [Context Compression](https://docs.docsgpt.cloud/Guides/compression): Reduce context while preserving key information. - [OCR for Sources and Attachments](https://docs.docsgpt.cloud/Guides/ocr): OCR behavior for ingestion and chat uploads. - [How to Use Different LLMs](https://docs.docsgpt.cloud/Guides/How-to-use-different-LLM): Additional model-selection guidance. - [Avoiding Hallucinations](https://docs.docsgpt.cloud/Guides/My-AI-answers-questions-using-external-knowledge): Improve answer grounding with external knowledge. - [Amazon Lightsail Deployment](https://docs.docsgpt.cloud/Deploying/Amazon-Lightsail): Deploy DocsGPT on AWS Lightsail. - [Railway Deployment](https://docs.docsgpt.cloud/Deploying/Railway): Deploy DocsGPT on Railway. - [Changelog](https://docs.docsgpt.cloud/changelog): Project release history.