Integrate MCP clients with the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Tools
retrieveContent
Retrieve content from Graphlit knowledge base. Accepts a search prompt, optional recency filter (defaults to last 30 days), and optional content type and file type filters. Prompt should be optimized for vector search, via text embeddings. Rewrite prompt as appropriate for higher relevance to search results. Returns the content sources in XML format, including metadata and Markdown text.
queryFeeds
Query feeds by name and/or feed type. Returns the feed name, identifier and other feed properties.
deleteFeed
Delete a feed and all of its ingested content. Accepts a feed identifier which was returned from one of the ingestion tools, like ingestGoogleDriveFiles. Content deletion will happen asynchronously. Returns the feed identifier and feed state, i.e. Deleted.
isFeedDone
Check if an asynchronous feed has completed ingesting all the available content. Accepts a feed identifier which was returned from one of the ingestion tools, like ingestGoogleDriveFiles. Returns whether the feed is done or not.