{"$schema":"https://wellknown.network/schemas/agent-record-v1.json","schemaVersion":"1","id":"ag_76r9nh4pfnmq","handle":"screenscribe-mcp","url":"https://wellknown.network/agents/screenscribe-mcp","links":{"self":"https://wellknown.network/agents/screenscribe-mcp/record.json","html":"https://wellknown.network/agents/screenscribe-mcp","markdown":"https://wellknown.network/agents/screenscribe-mcp/record.md","api":"https://wellknown.network/api/v1/agents/screenscribe-mcp","status":"https://wellknown.network/api/v1/agents/screenscribe-mcp/status","claim":"https://wellknown.network/agents/screenscribe-mcp/claim","claimApi":"https://wellknown.network/api/v1/claims","claimDescriptor":"https://wellknown.network/agents/screenscribe-mcp/claim.json","badge":"https://wellknown.network/agents/screenscribe-mcp/badge.svg","openapi":"https://wellknown.network/openapi.json","history":"https://wellknown.network/api/v1/agents/screenscribe-mcp/history","tools":"https://wellknown.network/api/v1/agents/screenscribe-mcp/tools"},"ard":{"identifier":"urn:air::server:screenscribe-mcp","type":"application/mcp-server-card+json"},"kind":"mcp_server","declared":{"name":"screenscribe-mcp","summary":"Extract typed, schema-validated data from videos and synthesize across many — a Gemini-powered CLI and MCP server for agents.","description":"# screenscribe\n\n**Extract *typed* data from videos — and synthesize across many.**\n\nThe best how-to knowledge — live demos, conference talks, tutorials, cooking videos — is trapped in video: you can watch it, but you can't *use* it. screenscribe turns it into structured, validated data an agent can act on. Hand it a schema and it fills it from what's on screen *and* said; point it at a whole channel and it synthesizes one coherent artifact across every video.\n\n**Powered by Gemini (it watches the video); your agent does the reasoning.** Standalone CLI + MCP server (Claude Code, Cursor, Windsurf, any MCP client). One key — and transcript-only mode needs none. Works regardless of the video's language and needs no captions.\n\n---\n\n## What it does — three layers\n\n1. **Watch & extract.** Gemini watches the video. Pull the transcript (free), a cheap whole-video analysis, or the key frames as PNG images your agent opens and reads.\n2. **Typed extraction** ⭐ — hand it a **JSON Schema** (or a bundled preset) and get back **validated JSON**: every CLI command shown, the final config file, a recipe with quantities, the code at each step. Prose is for humans to read; this is what an agent automates on.\n3. **Cross-video synthesis** ⭐ — point at a **channel / playlist / list of URLs** → fan the typed extraction out over every video → **compound** the results into one artifact conforming to an *aggregate* schema (a cookbook, a technique grammar, a comparison). It scales past a single context window, one capped batch at a time.\n\nThe raw extraction is a commodity (it's Gemini under the hood). The durable value is the **structure** and the **cross-video synthesis** — and both only get better as the models do.\n\n---\n\n## Why it matters\n\n- **Watch → automate.** *\"Extract every command shown in this setup tutorial\"* → a typed list your tooling runs. *\"Turn mom's whole cooking channel into one cookbook\"* → a synthesized artifact, not 300 summaries.\n- **Typed, not prose.** Output validates ag…","publisher":{"name":"ashmitb95","url":null},"homepage":"https://github.com/ashmitb95/screenscribe-mcp","repository":"https://github.com/ashmitb95/screenscribe-mcp","version":"0.4.0","license":null,"protocols":["mcp"],"tags":["agents","ai","cross-video-synthesis","gemini","json-schema","mcp","model-context-protocol","structured-extraction","transcript","video","youtube"],"pricing":null,"endpoints":[{"url":"pypi:screenscribe-mcp","type":"package_pypi","auth":null,"probeable":false}],"skills":null,"tools":null,"extra":null,"attribution":{"kind":"pypi","name":"pypi","repoUrl":"pypi","summary":"pypi","version":"pypi","description":"pypi","homepageUrl":"pypi","publisherName":"pypi"}},"derived":{"capabilities":[{"slug":"documents.extraction","name":"Data Extraction","confidence":1,"provenance":"derived"},{"slug":"research.web","name":"Deep Web Research","confidence":1,"provenance":"derived"}],"categories":["documents","research"],"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":"screenscribe-mcp","registry":"pypi","observedAt":"2026-10-28T22:24:21.837Z","publishedAt":"2026-07-02T10:15:24.071961Z","latestVersion":"0.4.0"},"toolSurface":null,"endpointFacts":[]},"verification":{"claimed":false,"claimedAt":null,"proofs":[]},"provenance":{"sources":[{"source":"pypi","key":"screenscribe-mcp","url":"https://pypi.org/project/screenscribe-mcp/","firstSeenAt":"2026-09-10T12:21:04.164Z","fetchedAt":"2026-10-28T22:22:21.490Z","normalizedAt":"2026-10-28T22:22:21.490Z"}]},"firstSeenAt":"2026-09-10T12:21:04.164Z","updatedAt":"2026-10-28T22:24:21.837Z"}