{"_id":"@dpm-tools/mcp-yt-clip","name":"@dpm-tools/mcp-yt-clip","dist-tags":{"latest":"0.1.0"},"versions":{"0.1.0":{"name":"@dpm-tools/mcp-yt-clip","version":"0.1.0","description":"MCP server: fetch YouTube transcripts, search for moments by keyword, return text-only summaries. Powered by youtube-transcript. Free, MIT.","type":"module","main":"src/index.ts","bin":{"mcp-yt-clip":"src/index.ts"},"scripts":{"start":"node --experimental-strip-types src/index.ts","inspect":"npx @modelcontextprotocol/inspector node --experimental-strip-types src/index.ts"},"engines":{"node":">=22.18.0"},"keywords":["mcp","mcp-server","model-context-protocol","claude","youtube","transcript","captions"],"license":"MIT","author":{"name":"dpm","url":"digital product mill"},"repository":{"type":"git","url":"git+https://github.com/jeffereychown-Max/mcp-yt-clip.git"},"publishConfig":{"access":"public"},"dependencies":{"@modelcontextprotocol/sdk":"^1.0.0","youtube-transcript":"^1.2.1"},"gitHead":"ed371d797b70cc7cbc61294af44f53be8c7b8a26","_id":"@dpm-tools/mcp-yt-clip@0.1.0","bugs":{"url":"https://github.com/jeffereychown-Max/mcp-yt-clip/issues"},"homepage":"https://github.com/jeffereychown-Max/mcp-yt-clip#readme","_nodeVersion":"25.8.1","_npmVersion":"11.11.0","dist":{"integrity":"sha512-/4+wy8BQeHTHuYXwv2uGssgkZF6YW8QHccVEjSo7xQQ87rPSFviv2bdIiHNjeYVZNNXyix9YENwIsCCwf5gNKg==","shasum":"7da0245956fb556aceb76d486c4e1767cf01a096","tarball":"https://registry.npmjs.org/@dpm-tools/mcp-yt-clip/-/mcp-yt-clip-0.1.0.tgz","fileCount":7,"unpackedSize":14888,"signatures":[{"keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U","sig":"MEYCIQDdJkYIoe6a0KC2yP7YXK5Po3c8wPGpHxDfQeCs06qhJQIhALXBEVQvde1keccy3MK2R7gRfmuyxmt5zIwjmCl606I6"}]},"_npmUser":{"name":"dpm-tools","email":"williamjong19@gmail.com"},"directories":{},"maintainers":[{"name":"dpm-tools","email":"williamjong19@gmail.com"}],"_npmOperationalInternal":{"host":"s3://npm-registry-packages-npm-production","tmp":"tmp/mcp-yt-clip_0.1.0_1778945284635_0.8317262340007523"},"_hasShrinkwrap":false}},"time":{"created":"2026-05-16T15:28:04.505Z","0.1.0":"2026-05-16T15:28:04.781Z","modified":"2026-05-16T15:28:05.073Z"},"maintainers":[{"name":"dpm-tools","email":"williamjong19@gmail.com"}],"description":"MCP server: fetch YouTube transcripts, search for moments by keyword, return text-only summaries. Powered by youtube-transcript. Free, MIT.","homepage":"https://github.com/jeffereychown-Max/mcp-yt-clip#readme","keywords":["mcp","mcp-server","model-context-protocol","claude","youtube","transcript","captions"],"repository":{"type":"git","url":"git+https://github.com/jeffereychown-Max/mcp-yt-clip.git"},"author":{"name":"dpm","url":"digital product mill"},"bugs":{"url":"https://github.com/jeffereychown-Max/mcp-yt-clip/issues"},"license":"MIT","readme":"# mcp-yt-clip\n\n> 5 YouTube transcript tools. Fetch, search by keyword, get clip URLs at timestamps. Free, MIT.\n\nFor researchers, video summarizers, and anyone who doesn't want to watch 60 minutes for one key moment.\n\n---\n\n## Install\n\n```json\n{\n  \"mcpServers\": {\n    \"yt\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"@dpm-tools/mcp-yt-clip\"]\n    }\n  }\n}\n```\n\nRequires Node.js 22.18+.\n\n---\n\n## Tools (5)\n\n- **`get_transcript`** — Full transcript as `{text, start, duration, timestamp}` segments.\n- **`transcript_text`** — Just text (no timestamps) — ideal for LLM summarization.\n- **`search_transcript`** — Find moments by keyword. Returns matches with timestamps + clickable clip URLs.\n- **`clip_url_at_timestamp`** — Build a `youtube.com/watch?v=...&t=Ns` URL.\n- **`extract_video_id`** — Pull 11-char ID from any YouTube URL format.\n\n---\n\n## Example prompts\n\n- *\"Get the transcript of https://www.youtube.com/watch?v=dQw4w9WgXcQ\"*\n- *\"Summarize this video: <url>\"* (use transcript_text)\n- *\"Find every mention of 'AI safety' in this 2-hour talk: <url>\"*\n- *\"What URL goes straight to 12:34 in this video?\"*\n\nSee [`examples/prompts.md`](./examples/prompts.md) for more.\n\n---\n\n## Limitations\n\nPowered by [`youtube-transcript`](https://www.npmjs.com/package/youtube-transcript) which scrapes YouTube. **YouTube can rate-limit or change formats** — if a fetch fails, retry later or check video has captions enabled.\n\nFor long videos (3hr+), `transcript_text` may return large outputs. Use `search_transcript` to find specific moments instead.\n\n---\n\n## Sister servers from dpm\n\n- **`mcp-devkit`**, **`mcp-public-data`**, **`mcp-diff`**, **`mcp-archive`**, **`mcp-citations`**, **`mcp-ical`**, **`mcp-dict`**, **`mcp-http-inspect`**, **`mcp-morning-brief`**, **`mcp-pantry`**.\n\n---\n\n## License\n\nMIT © dpm (digital product mill)\n","readmeFilename":"README.md","_rev":"1-f10eb7c6239631486b0311516f67e719"}