A
licenseNot graded
qualityB
maintenanceEnables AI coding agents to capture screen and voice recordings, extract timestamped frames, and receive structured Markdown reports with context for bug fixing and UI feedback.
9 npm
18
MIT
talkthrough-mcp
by korovin-aa97
Related Servers
Alternatives to talkthrough-mcp
No user-submitted related servers found.
Related Servers
- AlicenseAqualityBmaintenanceTurn screen recordings into automated Playwright E2E reproduction tests, structured GitHub bug reports with clickable video timestamps, and SOP walkthroughs.611 npmMIT
- AlicenseNot gradedqualityAmaintenanceAn MCP server that watches for screen recordings, transcribes spoken feedback locally, and attaches each comment to the relevant frame, enabling coding agents to see and fix the issues you mention.5MIT
- AlicenseAqualityCmaintenanceVisual feedback as agent work packets: stakeholders pin on your live app and your AI coding agent reads each pin (selector, screenshot, DOM, thread, acceptance criteria) via MCP and ships the fix.391,174 npm1MIT
- AlicenseAqualityAmaintenanceRead visual bug reports from Feedbug in your AI coding agent: screenshot, console logs, failed network requests, session replay and the DOM context of the element the tester clicked. Trace the bug to its source file, comment and resolve without leaving the editor.711 npmMIT
- AlicenseNot gradedqualityAmaintenanceGive AI agents visual + structured feedback about live web UIs — and give humans deterministic demo reels while you're at it.125 PyPI3Apache 2.0
TDQS
A4.9/5.0
Scored across 9 tools
Disambiguation5/5
Each tool targets a distinct stage of the pipeline: ingestion (process_url vs process_media), browsing (list_jobs), retrieval (get_transcript, get_frames, extract_frame, get_moment), search, and annotation (label_speakers). The descriptions include explicit 'When NOT to use' sections and anti-examples that sharply delineate boundaries, e.g. get_frames (stored keyframes) vs extract_frame (exact native-resolution frame) vs get_moment (bundled evidence).
Naming Consistency5/5
Names are uniformly snake_case and follow a predictable verb_noun pattern (process_url, process_media, list_jobs, extract_frame, get_frames, get_moment, get_transcript, label_speakers). The single tool 'search' is a minor stylistic deviation but remains natural and readable within the set.
Tool Count5/5
Nine tools is well-scoped for a media ingestion and analysis server. Each tool earns its place by covering a required capability without redundancy or bloat.
Completeness4/5
The surface covers ingestion (local and URL), listing, transcript retrieval, frame access, evidence bundling, lexical search, and speaker labeling—a complete workflow. The one notable gap is a delete/GC tool for managing stored jobs; the server only mentions a CLI `talkthrough-mcp gc`, leaving no in-band way to remove recordings.
Maintenance
ActivityActive
ResponsivenessWithin a week