Channel Brains
Indexes publicly available YouTube captions into a searchable local SQLite database, providing tools to create channel brains, search caption chunks with timestamped citations back to the original videos, and retrieve video transcripts.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Channel BrainsCreate a brain for https://www.youtube.com/@OpenAI"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Channel Brains
Channel Brains is a local, stdio MCP server that indexes publicly available YouTube captions into a searchable SQLite FTS5 database. It runs on your computer, needs no API key, and preserves timestamped links back to the original videos.
It is deliberately small and local:
No hosted service, Docker, browser dashboard, embeddings, or LLM runtime
No YouTube API key, database server, or cloud account
One local SQLite database per user, protected by a cross-process ingestion lock
One channel ingestion job at a time, with resumable per-video progress
Search results cite the original video and timestamp
Install
Requires uv.
After the public v0.1.0 release, run it directly from GitHub:
uvx --from "git+https://github.com/Pu11en/channel-brains@v0.1.0" channel-brains-mcpThe release is published from Pu11en/channel-brains.
For development:
uv sync --extra dev
uv run channel-brains-mcpThe server communicates only through standard input and output. Do not run it as an HTTP service.
Related MCP server: youtube-mcp
First workflow
Add the server to an MCP client using one of the configurations below.
Call
create_brainwith a supported YouTube channel URL, such ashttps://www.youtube.com/@OpenAI.Poll
get_brain_statusuntil the brain isready,paused, orfailed.Use
search_brainto retrieve timestamped caption matches.
The initial ingestion scans the complete channel listing so it can select up to 50 videos by view count. Caption availability and YouTube rate limits determine how much can be indexed.
MCP tools
Channel Brains exposes exactly these six tools:
Tool | Purpose |
| Validate a channel URL, persist a brain, and start local ingestion. |
| Read progress, counts, selection method, and errors without network access. |
| Page through indexed, skipped, pending, and failed video records. |
| Search local FTS5 caption chunks and return timestamped YouTube citations. |
| Page through stored caption chunks for one indexed video. |
| Remove a completed brain and all its local records. Active ingestion is refused. |
Local data and privacy
By default, Channel Brains stores its SQLite database and lock file in the platform’s user-data directory:
Windows:
%LOCALAPPDATA%\channel-brainsmacOS:
~/Library/Application Support/channel-brainsLinux:
~/.local/share/channel-brains
Set CHANNEL_BRAINS_HOME before launching the MCP server to use another location:
CHANNEL_BRAINS_HOME=/path/to/channel-brains-data uvx --from "git+https://github.com/Pu11en/channel-brains@v0.1.0" channel-brains-mcpCaptions and search indexes stay on the local machine. The only network requests are public YouTube requests made by yt-dlp and caption URL retrieval during ingestion.
MCP client configuration
All examples use the published Pu11en/channel-brains release command.
Hermes Agent
Add this entry under mcp_servers in your Hermes configuration, then restart Hermes:
mcp_servers:
channel_brains:
command: uvx
args:
- --from
- git+https://github.com/Pu11en/channel-brains@v0.1.0
- channel-brains-mcp
timeout: 120
connect_timeout: 60Hermes discovers the six tools at startup and registers them with an mcp_channel_brains_ prefix.
Claude Code
Register it at user scope:
claude mcp add -s user channel-brains -- \
uvx --from "git+https://github.com/Pu11en/channel-brains@v0.1.0" channel-brains-mcpConfirm it is available:
claude mcp listCodex CLI
Add this to ~/.codex/config.toml:
[mcp_servers.channel-brains]
command = "uvx"
args = ["--from", "git+https://github.com/Pu11en/channel-brains@v0.1.0", "channel-brains-mcp"]Restart Codex after saving the file.
OpenCode
Add a local MCP server entry to your OpenCode configuration:
{
"$schema": "https://opencode.ai/config.json",
"mcp": {
"channel-brains": {
"type": "local",
"command": [
"uvx",
"--from",
"git+https://github.com/Pu11en/channel-brains@v0.1.0",
"channel-brains-mcp"
],
"enabled": true
}
}
}Restart OpenCode to load the server.
Development and verification
uv sync --extra dev --locked
uv run ruff check .
uv run pytest
uv buildThe test suite is offline except for the deliberately separate manual live YouTube smoke test.
Limitations
Public YouTube captions can be unavailable, expired, restricted, or rate-limited.
Search is lexical SQLite FTS5 search, not semantic search or an answer-generation system.
The server indexes captions only. It does not download videos, reuse video footage, or create a knowledge graph.
A channel can contain many videos. The first local ingestion may take time.
License
MIT. See LICENSE.
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Servers
- AlicenseAqualityBmaintenanceAn MCP server (stdio + HTTP/SSE) that fetches video transcripts/subtitles via yt-dlp, with pagination for large responses. Supports YouTube, Twitter/X, Instagram, TikTok, Twitch, Vimeo, Facebook, Bilibili, VK, Dailymotion. Whisper fallback — transcribes audio when subtitles are unavailable (local or OpenAI API). Works with Cursor and other MCP hostLast updated817MIT
- Alicense-qualityDmaintenanceA local MCP server for extracting YouTube video transcripts, metadata, and performing visual analysis using Gemini Vision or local Whisper models. It enables users to process video content through various tools for subtitle retrieval and frame analysis.Last updated29MIT
- Alicense-qualityCmaintenanceMCP server for fetching YouTube video transcripts without an API key.Last updatedGPL 3.0
- Alicense-qualityDmaintenanceA minimalist MCP server that fetches transcripts from YouTube videos using the Supadata API.Last updated4MIT
Related MCP Connectors
Search your AI chat history (ChatGPT, Claude, Codex) from any MCP client. Remote, private, read-only
Any social-video URL → transcript, metadata, frames, OCR, summary, search, Q&A. MCP server + x402.
Local-first RAG engine with MCP server for AI agent integration.
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/Pu11en/channel-brains'
If you have feedback or need assistance with the MCP directory API, please join our Discord server