The BrainGraph MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your BrainGraph account. It provides structured and secure access to your knowledge brains, source documents, and graph facts, so your agent can ingest content, track processing, search facts, resolve entities, and find recurring patterns on your behalf.
- Content ingestion: Have your agent add public web pages, YouTube videos, audio recordings, document files, or plain text to your knowledge graph.
- Document tracking and review: Let the agent list your documents, filter them by name, notes, brain, or failed status, and check their transcription, processing, and analysis progress.
- Knowledge graph search: Instruct your agent to ask natural-language questions and retrieve relevant facts from all or selected knowledge brains.
- Entity resolution: Direct your agent to match a person, organization, place, or other named entity to its canonical record in BrainGraph.
- Relationship pattern discovery: Have your agent find recurring chains of relationships across your knowledge graph and focus on patterns that appear a chosen number of times.