Local Deep Research
github.com
- Free plan
- Windows
- macOS
- Linux
- Web
- API
- Self-hosted
8.0 Club score Visit github.com
What the maker publishes
- API and assistant integration
- It provides authenticated HTTP and Python APIs and an MCP server for Claude Desktop and Claude Code. source
- Data security
- The project says each user gets an isolated SQLCipher database encrypted with AES-256 using a key derived from their password, and that passwords are not stored. source
- Deployment
- Installation options include Docker, Docker Compose and pip; the README says pip installation works on Windows, macOS and Linux. source
- Deployment
- Installation options include Docker Compose, Docker, pip, and an Unraid template; the installation guide names CPU-only Compose support for all platforms and Windows PowerShell instructions. source
- Egress limitation
- The project describes egress scopes as best-effort guard rails rather than an air gap or hard security boundary. source
- Egress limits
- The project describes egress scopes as best-effort guard rails rather than an air gap or hard security boundary. source
- Exports and subscriptions
- It supports PDF and Markdown exports and scheduled research digests delivered as Markdown reports or structured summaries. source
- Interfaces
- The project provides a web UI, Python API, authenticated HTTP REST API, and an MCP server for Claude Desktop and Claude Code. source
- Knowledge base
- It can download sources into an encrypted library, index them and search them alongside the live web. source
- Knowledge base
- It can index downloaded papers and web pages in an encrypted library so users can search those sources alongside the live web. source
- Knowledge base integrations
- LangChain-compatible retrievers can connect sources including FAISS, Chroma, Pinecone, Weaviate and Elasticsearch. source
- Knowledge integrations
- LangChain-compatible retrievers are supported, including FAISS, Chroma, Pinecone, Weaviate, and Elasticsearch. source
- License
- The repository states that Local Deep Research is distributed under the MIT License. source
- LLM integrations
- It supports local models through Ollama, LM Studio and llama.cpp, plus OpenAI, Anthropic Claude, Google Gemini, OpenRouter, Requesty and compatible endpoints. source
- LLM integrations
- Supported model options include Ollama, LM Studio, llama.cpp, OpenAI, Anthropic Claude, Google Gemini, OpenRouter, Requesty, and compatible custom endpoints. source
- MCP limitation
- The MCP server is intended for local STDIO use and has no built-in authentication or rate limiting. source
- MCP security limit
- The MCP server is intended for local STDIO use and has no built-in authentication or rate limiting. source
- Privacy
- The project says it has no telemetry, analytics, tracking, crash reporting or phone-home calls; network requests are for user-initiated searches, model calls and configured notifications. source
- Privacy
- The project says it has no telemetry, analytics, tracking, or crash reporting, and that network calls are initiated by the user’s configured search, model, or notification services. source
- Purpose
- Local Deep Research is an AI research assistant that performs agentic research with multiple LLMs and search engines, then produces cited reports. source
- Purpose
- Local Deep Research is an AI-powered research assistant that searches multiple sources and synthesizes cited reports. source
- Research modes
- It offers quick summaries, detailed research, report generation and document analysis. source
- Research modes
- It offers quick summaries, detailed research, report generation, and AI analysis of private documents. source
- Search integrations
- Listed sources include arXiv, PubMed, Semantic Scholar, Wikipedia, SearXNG, GitHub, Elasticsearch, Wayback Machine, The Guardian and Wikinews. source
Checked against the maker's pages on 5 Oct 2026.