qybaihe / mu
mu (μ): most of what a coding agent decides isn't code. mu hands 40 kinds of yes/no calls to a small model: fewer wasted tokens, fewer approvals on autopilot, fewer runs that stop halfway. Built on pi.
Why this job: says: coding agent; topics: coding-agent, ai-coding, agent-harness
What it does
Does this output matter? Is this command safe for the task? Is the work really done? A coding agent answers questions like these hundreds of times a session. Handing them to the main model or to fixed rules costs you in three ways:
Get started
npm i -g mu-agent
# Node.js 22.19 or newer
mu setup
# connect a model
mu
# start a session in the current directory- Maintenance
- Active
- Last commit yesterday
- Stars
- 588
- +16 this week
- Licence
- MIT
- Allows business use
- Downloads for
- 3 systems
- Mac, Windows, Linux
- Latest version
- v0.3.9
- Released 2 days ago
- Built with
- TypeScript
- 93% of the code
Is it looked after?
- Contributors
- 2
- Last 90 days
- Issues
- 80% closed
- 3 still open
- Pull requests
- 366 merged
- 2 waiting
- Releases
- 36
- 30+ in the last year
Similar repos
| Repo | Stars | Last commit | Licence | Downloads for |
|---|---|---|---|---|
| NeuralNomadsAI / CodeNomad | 2,663 | yesterday | MIT | Mac, Windows, Linux |
| HarnessMD / munder-difflin | 9,454 | 3 days ago | MIT | Mac, Windows, Linux |
| OpenCoworkAI / open-cowork | 2,267 | 21 days ago | MIT | Mac, Windows |
More: download files, languages, topics and project record
Download files, v0.3.9
- Mac2 files, 10 downloads
- Windows2 files, 17 downloads
- Linux4 files, 11 downloads
Languages
- TypeScript 93%
- JavaScript 3%
- HTML 2%
- CSS 1%
Topics
- agent-harness
- ai-agent
- ai-coding
- cli
- coding-agent
- context-engineering
- desktop-app
- developer-tools
- electron
- jev
- llm
- multi-agent
Project record
- JobCoding agents
- StatusListed
- Commits8,740
- Forks64
- Watching0
- Stars in 28 days+86
- CreatedSep 22, 2026
- First seen by usNov 7, 2026
- CheckedNov 10, 2026