#8787th of 118 in Agent frameworks and SDKs
skill-recorder
Electron app that turns a screen recording into an agent skill
- Stars
- 5.7k
- License
- MIT
- Last commit
- Nov 2026
- Last release
- Nov 2026
- Language
- TypeScript
8 on 28 Jul 20265,748 on 10 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Skill Recorder captures a screen session (window switches, browser URLs, clipboard previews, optional recorded terminal and narration), or imports an MP4/WebM file. GitHub Copilot CLI then reconstructs an intent and ordered steps, from which it generates a SKILL.md or a scheduled automation that prefers native tools over UI replay. Capture and Whisper transcription run locally; analysis sends data to GitHub's cloud.
Who it is for: Agent users who want to teach agents repeatable tasks by demonstration
Strengths
- Generates SKILL.md files or scheduled automations from a single recorded demonstration
- Narration is transcribed on-device with Whisper, covering 99 languages
- Imports MP4/WebM recordings, so no live capture or screen permission is needed
- Fixture-based eval suite for the describer and builders (npm run eval)
Weaknesses
- Analysis requires a GitHub account with Copilot access; no other model provider is documented
- Analyze uploads frames, URLs, clipboard previews and terminal output to GitHub's cloud
- Source release only: installer builds the app locally and takes several minutes
- Import limited to 512 MiB, 15 minutes and 4K; peak memory about 4.6 GB measured
What it needs
- no GPU
- Needs GitHub Copilot CLI 1.0.78 (bundled), GitHub account with Copilot access, Node.js 24 (development only)
- Models: GitHub Copilot, Whisper (local transcription)
Quick answers
Is skill-recorder open source?
Yes. Its code is on GitHub under the MIT license.
Does skill-recorder need a GPU?
No, it runs without one.
Also in Agent frameworks and SDKs
See all 118| Rank | Project | Score |
|---|---|---|
| 1 | ppt-masterAgent skill that turns documents into natively editable PowerPoint decks | 95 out of 100 |
| 2 | openai-agents-pythonPython SDK for multi-agent workflows with handoffs, guardrails and tracing | 91 out of 100 |
| 3 | mastraTypeScript framework for building AI agents, workflows and MCP servers | 91 out of 100 |
| #4 | browser-usePython library that lets an LLM agent control a web browser | 90 out of 100 |
| #5 | copilotkitSDK for adding agent chat, generative UI and shared state to apps | 90 out of 100 |