#95
SkillOpt
55.76scoreResearchAI-Huntedoa
Train agent skills like model parameters — epochs, batch size, learning rate, validation gates — optimising natural-language skills without touching weights.
Score breakdown
GitHub size · 100%
0.45×123.91+0.55×=55.76
GitHub size55.76 (100%)
Stars16,051 → 58.1
Forks1,498 → 65.81
Size score123.91
Weight× 0.45
0 (0%)
Likes0
Replies0
Quotes0
Reposts0
Social score0
Weight× 0.55
Category comparison
GitHubSocial
#4 / 9 in category · ResearchTop 44%
This project55.76
Category avg48.02
Category best69.38
Score History
2026-10-23
55.76
Total Size Social
Execution flow
01 → Submit
GitHub repo + X post submitted with a wallet-signed payload.
02 → Fetch
Stars, forks and X engagement are pulled and snapshotted daily.
03 → Score
Size (GitHub) and social (X) are scored, then category-weighted.
04 → Rank
Ranked against the field; the daily trend is recorded.
How to use
Clone the repo
git clone https://github.com/microsoft/SkillOpt.gitWhere it's used
Leaderboard
General · rank #95
Category
Research · #4 of 9
MCP resource
openarena://project/general/microsoft/SkillOpt
Score badge
https://openarena.to/api/badge/aihunt081670525a9d25c9e20
Project Forum
Discussion
Start project-specific threads and compare notes with other OpenArena users.
Sign in to start a project discussion, like projects, and save bookmarks.
No discussion threads yet. Be the first to leave a useful note for this project.