haljishi/vidwords-mcp
by Various
Hosted YouTube MCP server for AI agents — search transcripts, read a video's frames, and cite exact timestamps. Works with Claude, Claude Code, Cursor, Codex and ChatGPT.
MCP
haljishi/vidwords-mcp
Added 8 Sept 2026
Overview
Hosted YouTube MCP server for AI agents, letting them search video transcripts, read frames, and cite exact timestamps. It works with Claude, Claude Code, Cursor, Codex, and ChatGPT.
Best for
Best for
Developers building AI agents that need structured, time-coded access to YouTube video content
Use cases
- Search video transcripts for specific topics or phrases
- Extract frames for visual context in agent workflows
- Cite precise timestamps when referencing YouTube content
How to use
Tools exposed
search_transcriptget_transcriptlist_channel_videoslist_watchlistswatchlist_activityanalyze_videoget_analysisask_video
Tested with
Claude Desktop, Claude Code, Cursor, ChatGPT
Notes
Hosted YouTube MCP server for AI agents, letting them search video transcripts, read frames, and cite exact timestamps. It works with Claude, Claude Code, Cursor, Codex, and ChatGPT.
0 stars on GitHub. Last updated 2026-08-22. Licensed MIT.
Use cases
- Search video transcripts for specific topics or phrases
- Extract frames for visual context in agent workflows
- Cite precise timestamps when referencing YouTube content
Pros
- Hosted, so no local setup or YouTube API key needed
- Works with multiple major AI chat and coding clients
- Combines transcript, frame, and timestamp access in one server
Cons
- No community engagement yet (0 stars) suggests early or unproven stage
- Hosted service carries potential uptime and data-privacy tradeoffs
- JavaScript-only implementation may not fit all agent infrastructures
Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.
Pros
- Hosted, so no local setup or YouTube API key needed
- Works with multiple major AI chat and coding clients
- Combines transcript, frame, and timestamp access in one server
Cons
- No community engagement yet (0 stars) suggests early or unproven stage
- Hosted service carries potential uptime and data-privacy tradeoffs
- JavaScript-only implementation may not fit all agent infrastructures
Get the free Developer’s Field Guide
A 27-page field guide to the AI coding workflow with Claude. Claude Code, MCP servers, the prompt patterns that work, and what to delegate. Free.
Enter your work email. We send it straight over, plus a few short notes worth knowing. Unsubscribe any time.
