GitHub Blog · AI & ML· Andrea Griffiths·· 3 d ago
Build once, run on any agent: A practical guide to agent skills
Build once, run on any agent: A practical guide to agent skills
AI summary
Agent skills capture repeatable engineering work in files teams can review and reuse, while tools, project context, and permissions remain specific to the host that runs them.
Selection record
Threshold 60Official, first-handFirst 58Second 58
Not admittedSum of both 116 < twice the threshold 120
- Source tier
- Official, first-hand; this tier's threshold is 60
- Pre-filter
- passed：正文详述Agent技能构建与用法
A model scores each item twice, independently, against one written standard, out of 100. An item is admitted only when the two scores add up to twice the threshold. The threshold is set per source tier.
Source: GitHub Blog · AI & ML · github.blog