Hi! Before opening a PR I wanted to check fit and timing, since your guidelines say brand-new skills aren't accepted until they have real community usage.
Repo: https://github.com/timwukp/agent-skills-best-practice (35 skills, Agent Skills spec, MIT/Apache 2.0)
The skill I'd propose for the list is fsi-compliance-checker — maps code/architecture/infra changes to specific control IDs in PCI-DSS v4.0 and MAS TRM (Singapore financial regulator), with per-framework reference files loaded on demand:
https://github.com/timwukp/agent-skills-best-practice/tree/main/skills/skills/fsi-compliance-checker
Why it might be a useful addition: financial-services compliance is close to empty in the current list (I couldn't find any MAS TRM coverage), and the skill ships evals plus a documented 4-layer test methodology (blind trigger routing, independently graded task evals, real Claude Code verification — see the repo's TESTING.md).
The repo is young (low stars), so my question: would you prefer I wait until it shows more adoption, or is a niche-coverage skill with documented testing worth reviewing now? Happy to follow whatever bar you set — and to submit it under "Community Skills → Development and Testing" or "Other" per your format if/when appropriate.
Hi! Before opening a PR I wanted to check fit and timing, since your guidelines say brand-new skills aren't accepted until they have real community usage.
Repo: https://github.com/timwukp/agent-skills-best-practice (35 skills, Agent Skills spec, MIT/Apache 2.0)
The skill I'd propose for the list is fsi-compliance-checker — maps code/architecture/infra changes to specific control IDs in PCI-DSS v4.0 and MAS TRM (Singapore financial regulator), with per-framework reference files loaded on demand:
https://github.com/timwukp/agent-skills-best-practice/tree/main/skills/skills/fsi-compliance-checker
Why it might be a useful addition: financial-services compliance is close to empty in the current list (I couldn't find any MAS TRM coverage), and the skill ships evals plus a documented 4-layer test methodology (blind trigger routing, independently graded task evals, real Claude Code verification — see the repo's TESTING.md).
The repo is young (low stars), so my question: would you prefer I wait until it shows more adoption, or is a niche-coverage skill with documented testing worth reviewing now? Happy to follow whatever bar you set — and to submit it under "Community Skills → Development and Testing" or "Other" per your format if/when appropriate.