BrainGraph
BrainGraph builds a searchable knowledge graph from documents, recordings, web pages, and other source material for grounded AI workflows.
Integration
- Vendor
- BrainGraph
- Category
- other
- Auth
- API_KEY
- Tools
- 10
- Composio slug
brain_graph
Tools
- Get Document Progress
Get the current transcription, processing, and analysis status for a BrainGraph document. Poll until transcript.status is analyzed; analysis is null while processing is incomplete.
- Ingest Web Page
Fetch readable content from a public web page and queue it for BrainGraph ingestion. This creates persistent data and consumes upload and URL-submission quota. After a timeout or other ambiguous failure, check for the document before retryi
- Ingest YouTube Video
Submit a YouTube video for metadata, transcript, and speaker processing; the result may only confirm that processing was queued. This creates persistent data and consumes upload and URL-submission quota. After a timeout or incomplete stream
- List Brains
List the connected BrainGraph account's knowledge brains and identify the default brain for filtering or targeting other tools.
- List Documents
List BrainGraph documents newest first, optionally filtering by filename or notes, failed ingestion status, and brain. Returns one page at a time.
- List Graph Patterns
Find recurring relationship-chain patterns in the connected BrainGraph knowledge graph, subject to a minimum occurrence count.
- Resolve Entity
Resolve an entity name to a canonical BrainGraph record. If there is no single match, return an explicit ambiguous or not-found result rather than treating it as an HTTP error.
- Search Brain
Search facts in the connected account's BrainGraph knowledge graph using a natural-language question, optionally restricted to selected brains.
- Transcribe Audio
Upload an audio file or submit a direct public audio URL for diarized transcription and graph ingestion. This creates persistent data and consumes upload and transcription allowance. Do not retry automatically after an ambiguous failure bec
- Upload Document
Upload a document file supplied by the caller or plain text and queue it for BrainGraph ingestion. This creates persistent data and consumes the account's upload allowance. After a timeout or other ambiguous failure, check for the document
Setup
- Create or open a Switchy Project.
- Open Integrations and connect BrainGraph.
- Authorise through api_key. Switchy stores the connection against your org.
- Everyone in the Project can now @-mention the BrainGraph tools in chat.
What teammates see: by default, memories from BrainGraph are scoped to the Project (PROJECT visibility) - you can mark any memory PRIVATE or share it ORG-wide.
Works well with
Top models
Compatibility data appears once enough Projects have used this MCP together with a given model.