Agent-S vs skill-recorder
Agent-S is much bigger: 12.2k stars against 3.8k. Over the days we have tracked them Agent-S moved +3.9% and skill-recorder +224.4%, so skill-recorder is growing faster right now.
They split the axes: Agent-S leads on model freedom, skill-recorder on setup ease.
Stars and commit dates come from our own daily tracking. The six axes are read off each project's documentation by our review pipeline, so they describe what a project says about itself, not what we measured in its code.
Where they stand today
An open-source, research-backed GUI agent framework that combines grounding models and optional local code execution and achieves SOTA (surpassing human performance on OSWorld).
- Stars
- 12.2k
- Tracked growth
- +3.9%
- Maturity
- ●●●●●
- Last commit
- 35d ago
- Language
- Python
- License
- Apache-2.0
- Cost to run
- Your API keys (provider-dependent)
Converts a single on-screen recording (clicks, window switches, URLs, and optional narration) into a generalized, reusable Skill or scheduled Automation using the GitHub Copilot CLI.
- Stars
- 3.8k
- Tracked growth
- +224.4%
- Maturity
- ●●●●●
- Last commit
- 1d ago
- Language
- TypeScript
- License
- MIT
- Cost to run
- Requires GitHub Copilot access
Six axes, head to head
Each axis runs 0 to 5. The label under a score is what that project's own docs claim, not a category average.
| Axis | Agent-S | skill-recorder |
|---|---|---|
Context depth How much of your codebase it sees before it answers: the open diff, the diff plus related files, or the whole repository. | ●●●●● Single-screen context | ●●●●● Recording only |
Noise control How it keeps output volume down — severity thresholds, deduplication, incremental runs over new commits only. | ●●●●● Basic filters | ●●●●● Manual review/edit |
Customization How far it bends to your team: custom rules, prompts, style guides, per-path config. | ●●●●● Configurable params | ●●●●● Manual edits only |
Privacy Whether your code stays on your own infrastructure: fully local, self-hostable, or cloud API only. | ●●●●● Cloud APIs (your key) | ●●●●● Copilot cloud required |
Model freedom Whether you can point it at any provider, or it is wired to one. | ●●●●● Bring-your-own key | ●●●●● Copilot only |
Setup ease What it takes to get a first useful run out of it. | ●●●●● Config + API key | ●●●●● One-command install |
Which one to pick
Pick Agent-S if…
SOTA performance — pick Agent S when you need a research-grade, open-source GUI agent with grounding-model support and optional local code execution.
- Model freedom: Bring-your-own key (5/5 against 1/5)
Pick skill-recorder if…
Privacy-aware — captures and processes recordings locally and only uploads event data and images to GitHub Copilot when you explicitly choose Analyze, making it easy to create real-world reusable agent skills from one run.
- Setup ease: One-command install (5/5 against 3/5)
What people want from each one
simular-ai/Agent-S
Hacker News: Show HN: Agent S: an open agentic framework that uses computers drew 2 points and 1 comments.
microsoft/skill-recorder
Questions people ask
Is Agent-S better than skill-recorder?
They split the axes: Agent-S leads on model freedom, skill-recorder on setup ease. Agent-S is worth picking when sOTA performance — pick Agent S when you need a research-grade, open-source GUI agent with grounding-model support and optional local code execution.
Which of Agent-S and skill-recorder keeps my code private?
Agent-S: Cloud APIs (your key) (3/5). skill-recorder: Copilot cloud required (2/5).
What does each one cost to run?
Agent-S: Your API keys (provider-dependent). skill-recorder: Requires GitHub Copilot access.
Full profiles: simular-ai/Agent-S and microsoft/skill-recorder. Everything else in Computer use.