MCP server for LightRag
Connect Lightrag to Claude, Cursor or any other MCP client and it stops being a tab you switch to. MCP server for LightRag. The lightrag mcp server is what makes that connection.
Model Context Protocol server for seamless LightRAG integration.
The toolset is worth reading before you wire it up, because it tells you what the integration is really for:
insert_text — The insert_text tool exposed by this serverinsert_texts — The insert_texts tool exposed by this serverscan_documents — Trigger scanning process for new documents in the input directoryget_documents_paginated — The get_documents_paginated tool exposed by this serverget_document_status_counts — The get_document_status_counts tool exposed by this serverget_track_status — The get_track_status tool exposed by this serverdelete_document — The delete_document tool exposed by this serverclear_documents — The clear_documents tool exposed by this serverclear_cache — The clear_cache tool exposed by this serverreprocess_failed — 🔍 Queries (2 tools) - Search and retrieve informationquery_text — The query_text tool exposed by this serverquery_data — Advanced data retrieval endpoint for structured RAG analysis without LLM generationConfiguration is passed through the environment: LIGHTRAG_BASE_URL, LIGHTRAG_API_KEY. Treat anything key-shaped as a real credential — scope it to the minimum the server needs, and rotate it if it ever lands in a shared config.
The server ships on npm as @modelcontextprotocol/inspector, so your MCP client can launch it on demand — there is no separate build step. Add the server block to your client's configuration, restart it, and the tools register themselves.
Plenty of AI and media services servers cover similar ground. The differences that matter in practice are scope of access and how much setup stands between you and a working tool call. Lightrag's toolset — insert_text, insert_texts, scan_documents and 11 more — is a fair guide to whether it matches your workflow. It is maintained by deight; worth a glance at recent repository activity before you build anything load-bearing on it.
We check each listing at SyncDev against the project's documentation before it goes live — if something here drifts out of date, it is a bug worth reporting.
| Tool | What it does |
|---|---|
| insert_text | The insert_text tool exposed by this server. |
| insert_texts | The insert_texts tool exposed by this server. |
| scan_documents | Trigger scanning process for new documents in the input directory. |
| get_documents_paginated | The get_documents_paginated tool exposed by this server. |
| get_document_status_counts | The get_document_status_counts tool exposed by this server. |
| get_track_status | The get_track_status tool exposed by this server. |
| delete_document | The delete_document tool exposed by this server. |
| clear_documents | The clear_documents tool exposed by this server. |
| clear_cache | The clear_cache tool exposed by this server. |
| reprocess_failed | <details> <summary><b>🔍 Queries (2 tools)</b> - Search and retrieve information</summary> |
| query_text | The query_text tool exposed by this server. |
| query_data | Advanced data retrieval endpoint for structured RAG analysis without LLM generation. |
| get_graph_labels | The get_graph_labels tool exposed by this server. |
| get_popular_labels | Get popular labels by node degree (most connected entities). |
{
"mcpServers": {
"lightrag": {
"command": "npx",
"args": ["-y", "@modelcontextprotocol/inspector"],
"env": {
"LIGHTRAG_BASE_URL": "your-value",
"LIGHTRAG_API_KEY": "your-value"
}
}
}
}Add to claude_desktop_config.json, then restart Claude Desktop.
| Variable | Description | Required |
|---|---|---|
| LIGHTRAG_BASE_URL | Endpoint or connection string the server talks to. | Yes |
| LIGHTRAG_API_KEY | Credential the server authenticates with. | Yes |
Build a programmable telecommunications stack for connecting telephony services with the Internet via a cloud-based utility.
Search built for AI, not humans — semantic web search that returns model-ready content, plus code context.
Answers, not links — delegate questions to Perplexity's search-grounded models and get cited responses back.
Give your assistant a voice — text-to-speech, voice cloning and audio tools from the ElevenLabs API.
Give your assistant a real code sandbox — isolated cloud VMs for actually running the code it writes.
The ML hub in your context window — search models, datasets, papers and run Spaces from the official server.