braintube-mcp
# BrainTube MCP Server
**Save once. Query anywhere.** BrainTube compiles what you watch, read and listen to — YouTube videos, podcasts, articles, PDFs — into a persistent, searchable knowledge base, and this MCP server exposes it to every MCP-capable AI client over one endpoint.
- **Endpoint:** `https://mcp.brain-tube.com/mcp` (Streamable HTTP)
- **Version:** 3.12.4
- **Auth:** OAuth 2.0 (authorization code + PKCE) or a BrainTube API key via `X-BrainTube-Token` header
- **Discovery:** [`/.well-known/oauth-protected-resource`](https://mcp.brain-tube.com/.well-known/oauth-protected-resource) · [`/.well-known/oauth-authorization-server`](https://mcp.brain-tube.com/.well-known/oauth-authorization-server)
## What it does
Your AI can search and cite your own corpus instead of starting every conversation cold:
- **Semantic + keyword search** over everything you've saved, with citations back to the source (down to video timestamps)
- **Ingest** notes, articles and web content directly from any MCP client
- **Knowledge graph & related-item traversal** across your corpus
- **Session context** — expertise profile, recent activity, and resurfacing tools
- **Per-user isolation** — JWT-scoped access; you only ever see your own corpus
## Quickstart
1. Create an account at [brain-tube.com](https://brain-tube.com) and get an API key (or use OAuth from a compatible client).
2. Add the server to your MCP client:
```json
{
"mcpServers": {
"braintube": {
"type": "http",
"url": "https://mcp.brain-tube.com/mcp",
"headers": { "X-BrainTube-Token": "bt_..." }
}
}
}
```
Works with any MCP-capable client (Claude, Cursor, and others). In clients with OAuth support, just add the endpoint URL and sign in when prompted.
## Links
- Website: [brain-tube.com](https://brain-tube.com)
- MCP setup guide: [brain-tube.com/connect](https://brain-tube.com/connect)
- Official MCP registry: `io.github.lildaddyo/braintube-mcp`
- Support: [brain-tube.com/contact](https://brain-tube.com/contact)
TDQS
Scored across 28 tools
Most tools have distinct purposes, but the multiple search variants (deep_search, search_knowledge, search_by_date_range, search_by_source, search_obsidian) could cause confusion despite their specific scopes. The knowledge graph tools are well-differentiated.
All tool names follow a consistent snake_case convention with a clear verb_noun pattern. Examples like 'get_knowledge_graph', 'search_by_source', and 'sync_readwise' demonstrate uniform and predictable naming.
28 tools is on the higher side, but the broad scope of personal knowledge management—covering search, graph traversal, sync, health, and retrieval—justifies the count. Could potentially be trimmed by merging some search tools.
The server covers a comprehensive range of operations: search, knowledge graph, statistics, health, sync with Readwise, and brain interaction. Minor gaps exist, such as lack of manual item creation/deletion, but core workflows are well-supported.