Yao Meta Skill provides a Python-based framework to create, evaluate, package, and govern reusable agent skills, aiming to support end-to-end lifecycle governance and portability across targets. It emphasizes Skill OS 2.0 concepts and evidence-driven release processes.
Collecting history — the radar snapshots this repo daily. The trend line appears after 3 days of data (1 so far).
What it is
yao-meta-skill creates, evaluates, packages, and governs reusable agent skills. The 1.0 line focused on turning repeated workflows into installable, readable, cross-platform skill packages. The 2.0 line expands that factory into a Skill OS: a governed system for modeling a skill once, compiling it for multiple targets, testing its behavior, reviewing its release evidence, and tracking the next iteration.
How it works
Skill OS 2.0 introduces concepts such as Skill IR, target compilers/adapters, Output Eval Lab, Review Studio 2.0, evidence and release governance, and a SkillOps loop. The architecture maps inputs (workflow/prompts/transcripts) to an explicit intent model, then through Skill IR to a skill package and target-specific adapters, with evaluation, review, and release stages feeding back into the loop.
Getting started
The README provides command examples for operator diagnostics:
python3 scripts/yao.py install-status --expected-source .
python3 scripts/yao.py localized-doc-sync-check
python3 scripts/yao.py pr-review-report 4 --repo yaojingang/yao-meta-skill
Recent releases
Latest releases section lists: - none
Traction
Stars: 2209 Forks: 213 Open issues: 0 Created: 2026-03-31 Last push: 2026-07-16
Behind the repo
Not applicable (no linked startup/company section in the provided data).
Caveats
License: MIT Language: Python License and age facts are provided; no other caveats are listed in the provided material.