Hi, I'd like to suggest a complementary direction for hermes-agent-self-evolution.
Project:
https://github.com/AMAP-ML/SkillClaw
hermes-agent-self-evolution focuses on evolutionary self-improvement for Hermes. SkillClaw is adjacent, but narrower: long-lived skill libraries often become noisy, duplicated, stale, and fragmented across agents or devices.
Its role is a post-task skill evolution loop that deduplicates, merges, improves, and shares skills over time.
So the fit here is not replacing self-evolution, but acting as an externalized skill lifecycle / governance layer around repeated skill accumulation.
If useful, I can send a concise technical summary.
Paper:
https://arxiv.org/abs/2604.08377
Hi, I'd like to suggest a complementary direction for
hermes-agent-self-evolution.Project:
https://github.com/AMAP-ML/SkillClaw
hermes-agent-self-evolutionfocuses on evolutionary self-improvement for Hermes. SkillClaw is adjacent, but narrower: long-lived skill libraries often become noisy, duplicated, stale, and fragmented across agents or devices.Its role is a post-task skill evolution loop that deduplicates, merges, improves, and shares skills over time.
So the fit here is not replacing self-evolution, but acting as an externalized skill lifecycle / governance layer around repeated skill accumulation.
If useful, I can send a concise technical summary.
Paper:
https://arxiv.org/abs/2604.08377