DeepRepo is a Python library for performing Retrieval Augmented Generation (RAG) on local codebases, supporting multiple AI providers and integrating with MCP clients.
From the registry: Productivity-boosting RAG engine for codebases with multi-provider AI support and semantic search.
$ pip install deeprepo[mcp]Please install the `deeprepo` MCP server into my current AI client (that's you).
Required prerequisites (do these first if not already done):
- **Install DeepRepo with MCP extras** — Install the Python package with MCP dependencies. Run: `pip install deeprepo[mcp]`
- **Choose an AI provider** — Ollama (local, free), HuggingFace, OpenAI, Anthropic, or Gemini. Set the corresponding API key env var if needed.
Canonical MCP server config (stdio transport):
- command: `python`
- args: ["-m","deeprepo.mcp.server"]
- optional environment variables:
- `LLM_PROVIDER`: LLM provider: ollama, huggingface, openai, anthropic, gemini (example: `ollama`)
- `EMBEDDING_PROVIDER`: Embedding provider (can differ from LLM provider) (example: `openai`)
- `OPENAI_API_KEY`: OpenAI API key (example: `sk-<your_openai_key>`)
- `ANTHROPIC_API_KEY`: Anthropic API key (example: `sk-ant-<your_key>`)
- `HUGGINGFACE_API_KEY`: HuggingFace API token (example: `hf_<token>`)
- `GEMINI_API_KEY`: Google Gemini API key (example: `<gemini_key>`)
Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.OPENAI_API_KEYrequiredOpenAI API keyANTHROPIC_API_KEYrequiredAnthropic API keyHUGGINGFACE_API_KEYrequiredHuggingFace API tokenGEMINI_API_KEYrequiredGoogle Gemini API keyMCP server for searching Airweave collections with natural language queries.