{"$schema":"https://wellknown.network/schemas/agent-record-v1.json","schemaVersion":"1","id":"ag_qcs6m7k6c8q3","handle":"youty-mcp","url":"https://wellknown.network/agents/youty-mcp","links":{"self":"https://wellknown.network/agents/youty-mcp/record.json","html":"https://wellknown.network/agents/youty-mcp","markdown":"https://wellknown.network/agents/youty-mcp/record.md","api":"https://wellknown.network/api/v1/agents/youty-mcp","status":"https://wellknown.network/api/v1/agents/youty-mcp/status","claim":"https://wellknown.network/agents/youty-mcp/claim","claimApi":"https://wellknown.network/api/v1/claims","claimDescriptor":"https://wellknown.network/agents/youty-mcp/claim.json","badge":"https://wellknown.network/agents/youty-mcp/badge.svg","openapi":"https://wellknown.network/openapi.json"},"ard":{"identifier":"urn:air::server:youty-mcp","type":"application/mcp-server-card+json"},"kind":"mcp_server","declared":{"name":"youty-mcp","summary":"Youty MCP server — exposes the Youty vault index (sqlite-vec + FTS5) to MCP-compatible AIs.","description":"# youty-mcp\n\nLocal MCP server that exposes the Youty vault's vector index to any\nMCP-compatible AI (Claude Desktop, Claude Code, Cursor).\n\n## What it does\n\nSeven tools, hybrid dense + BM25 retrieval over your captured YouTube /\nInstagram / TikTok videos, plus joint text → frame retrieval via\nGoogle's SigLIP-Base-Patch16-224 (Apache-2.0). Search runs locally on macOS;\nlatency depends on model warm-up, hardware, and vault size.\n\n| Tool | Returns |\n|---|---|\n| `search(query, k=15, platform?, since_iso?)` | hybrid dense + BM25 + RRF over transcript chunks; top-k results with `frame` paths + `video_md_path` |\n| `search_frames(query, k=10, platform?)` | SigLIP-Base joint text→image; top-k frame matches with parent video metadata |\n| `get_transcript(video_id)` | full `video.md` + parsed frontmatter — the whole video into context |\n| `get_video(video_id)` | frontmatter + folder listing + frame paths |\n| `view_frames(video_id, frame_ms?, max_frames=6)` | the frame JPEGs themselves, as MCP image content — viewable in **any** client |\n| `list_videos(platform?, channel?, limit=100)` | newest-first listing |\n| `find_similar(video_id, k=10)` | nearest videos by averaged body-chunk vectors |\n\n**The loop:** `search` finds the relevant moments → `get_transcript` pulls the\nwords into context → `view_frames` loads the matching frames into the model's\nvision. `search` / `search_frames` also return raw frame *paths*, which require\na client with local-file access; `view_frames` returns the images inline, so\nthe visual half of the loop works in Claude Desktop, Cursor, and Claude Code\nalike.\n\n**On-screen text.** `search` covers both what a video *said* and what it\n*showed*: each result's `chunk.type` is `body`/`description`/`header` (spoken +\nmetadata) or **`frame_text`** — text recognized on-screen via on-device OCR\n(slides, code, terminal output, labels). A `frame_text` hit means the answer was\nvisible in the video; pair it with `view_frames` to see that moment.\n\n## Install\n\n```bash\ncd …","publisher":{"name":"Bent Eisheuer","url":null},"homepage":"https://github.com/legetdev/youty/blob/main/youty-mcp/README.md","repository":"https://github.com/legetdev/youty/blob/main/youty-mcp/README.md","version":"1.4.9","license":"MIT","protocols":["mcp"],"tags":["mcp","youty","vault","sqlite-vec","search","rag","youtube"],"pricing":null,"endpoints":[{"url":"pypi:youty-mcp","type":"package_pypi","auth":null,"probeable":false}],"skills":null,"tools":null,"extra":null,"attribution":{"kind":"pypi","name":"pypi","license":"pypi","repoUrl":"pypi","summary":"pypi","version":"pypi","description":"pypi","homepageUrl":"pypi","publisherName":"pypi"}},"derived":{"capabilities":[{"slug":"data.database","name":"Databases","confidence":1,"provenance":"derived"},{"slug":"security.secrets","name":"Secrets Management","confidence":1,"provenance":"declared"},{"slug":"data.vector-search","name":"Vector Search","confidence":1,"provenance":"declared"},{"slug":"content.social","name":"Social Media","confidence":0.814,"provenance":"derived"},{"slug":"infra.devices","name":"Devices & IoT","confidence":0.802,"provenance":"derived"},{"slug":"dev.terminal","name":"Terminal & Shell","confidence":0.779,"provenance":"derived"}],"categories":["content","data","dev","infra","security"],"language":"en"},"observed":{"status":"unknown","statusReason":"Distributed as a package to run locally; no network endpoint to check.","lastOkAt":null,"lastProbedAt":null,"statusComputedAt":null,"reliability30d":null,"latestObservations":[],"tools":null,"package":{"name":"youty-mcp","registry":"pypi","observedAt":"2026-09-10T16:24:29.741Z","publishedAt":"2026-09-07T21:38:06.558958Z","latestVersion":"1.4.9"}},"verification":{"claimed":false,"claimedAt":null,"proofs":[]},"provenance":{"sources":[{"source":"pypi","key":"youty-mcp","url":"https://pypi.org/project/youty-mcp/","firstSeenAt":"2026-09-10T16:23:32.569Z","fetchedAt":"2026-09-10T16:23:32.569Z","normalizedAt":"2026-09-10T16:23:32.569Z"}]},"firstSeenAt":"2026-09-10T16:23:32.569Z","updatedAt":"2026-09-10T16:24:29.741Z"}