scan_documents
Scan for new documents in LightRAG.
How to use it
scan_documents is exposed by the Daniel Lightrag MCP MCP server. Add the server to your MCP client (Claude Desktop, Cursor, Windsurf and others), and the scan_documents tool becomes available to the model automatically. See the full listing for setup details and every tool this server provides.
Install Daniel Lightrag MCP
pip install -e .Other tools in Daniel Lightrag MCP (35)
Check if an entity exists in the knowledge graph.
Clear LightRAG cache.
Clear all documents from LightRAG.
Delete a specific document by ID.
Delete an entity from the knowledge graph.
Delete a relation from the knowledge graph.
(required): ID of the document to delete
(required): ID of the entity to update
(required): Name of the entity to check
(required): Path to the file to upload
Get document status counts.
Retrieve all documents from LightRAG.
Retrieve documents with pagination.
Get labels from the knowledge graph.
Check LightRAG server health.
Retrieve the knowledge graph from LightRAG.
Get the pipeline status from LightRAG.
Get track status by ID.
Insert text content into LightRAG.
Insert multiple text documents into LightRAG.
(optional): Query mode - "naive", "local", "global", or "hybrid" (default: "hybrid")
(optional): Whether to only return context without generation (default: false)
(required): Page number (1-based)
(required): Number of documents per page (1-100)
(required): Properties to update
(required): Query text
Query LightRAG with text.
Stream query results from LightRAG.
(required): ID of the relation to update
(required): Text content to insert
(required): Array of text documents with optional title and metadata
(required): ID of the track to get status for
Update an entity in the knowledge graph.
Update a relation in the knowledge graph.
Upload a document file to LightRAG.