Hi, I'd like to suggest a complementary direction for agent-skills.
Project:
https://github.com/AMAP-ML/SkillClaw
Your project focuses on platform-specific agent skills and integrations.
SkillClaw focuses on a different but related problem that shows up once skill libraries grow over time:
- duplicated skills accumulate
- older skills become stale
- skills fragment across agents, devices, and workspaces
- long-term skill quality becomes harder to manage
Its role is a post-task skill evolution loop that:
- deduplicates overlapping skills
- merges related skills
- improves skill quality over time
- shares evolved skills across agents / devices / teams
I think it is complementary rather than competitive: your repo helps people create, discover, or use skills, while SkillClaw addresses skill lifecycle management after those skills have been used repeatedly.
If useful, I can put together a concise note, example workflow, or integration angle.
Paper:
https://arxiv.org/abs/2604.08377
Hi, I'd like to suggest a complementary direction for
agent-skills.Project:
https://github.com/AMAP-ML/SkillClaw
Your project focuses on platform-specific agent skills and integrations.
SkillClaw focuses on a different but related problem that shows up once skill libraries grow over time:
Its role is a post-task skill evolution loop that:
I think it is complementary rather than competitive: your repo helps people create, discover, or use skills, while SkillClaw addresses skill lifecycle management after those skills have been used repeatedly.
If useful, I can put together a concise note, example workflow, or integration angle.
Paper:
https://arxiv.org/abs/2604.08377