Talkthrough

Talkthrough is a Data skill for Claude Code, published by korovin-aa97 in talkthrough-mcp.

12 stars0 forkson korovin-aa97/talkthrough-mcpAdded 2026/07/16Repository updated 2026/07/17

agent-skillsai-agentsclaudedeveloper-toolsfeedbackffmpeglocal-firstmcpmcp-servermodel-context-protocolocrpythonscreen-recordingspeech-to-texttranscriptionvideo-analysiswhisper

Install in seconds

Install Talkthrough

Copy Talkthrough into your Claude Code skills folder. Run the command in your terminal, or review the source on GitHub before installing.

terminal

npx degit https://github.com/korovin-aa97/talkthrough-mcp/tree/main/.agents/skills/talkthrough ~/.claude/skills/talkthrough

Requires Node.js. Downloads this skill only — not the rest of the repository — into your Claude Code skills folder.

Without Node.js

git clone https://github.com/korovin-aa97/talkthrough-mcp.git

Clones the whole repository, then copy the skill’s own directory into your skills folder yourself.

In this catalog

Source file: .agents/skills/talkthrough/SKILL.md in korovin-aa97/talkthrough-mcp
Installs to: ~/.claude/skills/talkthrough
Collection: One of 2 skills cataloged from this repository
Category: Data — 812 skills

What Talkthrough does

Analyze narrated screen recordings and audio files to extract transcripts, timestamps, OCR text, and correlate spoken remarks with logs. Used to triage feedback, generate specs/backlogs, and extract action items from recordings.

Talkthrough is cataloged under Data on DirSkills. Talkthrough comes from a repository tagged agent-skills, ai-agents, claude, developer-tools and feedback.

Reviewed 2026-08-04

DirSkills review

This is the canonical DirSkills entry for Talkthrough; the repository ships the same skill text in two integration paths, so the duplicate listing now redirects here.

The workflow depends on a connected Talkthrough MCP server and is not a standalone transcription tool.
It correlates transcript, scene, OCR, and wall-clock evidence from narrated recordings for feedback triage and specification work.

Documentation

README

Excerpt

Analyzing narrated recordings with talkthrough

The talkthrough MCP server turns a local video/audio file into queryable structured data: timestamped transcript segments, scene keyframes, OCR'd on-screen text, and wall-clock anchoring. No LLM inside — you bring the reasoning; it brings the evidence. Everything is lazy and token-budgeted: never ask for more than the moment you are analyzing.

Prerequisite

The talkthrough MCP server must be connected (tools like process_media / get_transcript are visible). If not, tell the user to install it: claude mcp add -s user talkthrough -- uvx talkthrough-mcp (see the repository README for other clients).

This is the opening of the README. Read the full README on GitHub.

Frequently asked about Talkthrough

What else does korovin-aa97 publish alongside Talkthrough?
Talkthrough is one of 2 skills that DirSkills catalogs from korovin-aa97/talkthrough-mcp, the repository it ships in. Its siblings there include Talkthrough. Each one is a separate skill with its own page in this directory, installs the same way Talkthrough does, and is maintained by korovin-aa97 in that same repository. The rest of the collection is listed on the korovin-aa97/talkthrough-mcp page.
How does Talkthrough compare to other Data skills?
Talkthrough ranks #809 by stars among the 812 Data skills in this catalog. The most-starred ones next to it are Benchmark Methodology, Jupyter Notebook and Solana. DirSkills ranks by the star count of the repository each skill ships in, so that order reflects how popular those repositories are rather than any review of Talkthrough against them. Open each page to compare what they document and how they install.