A skill improves every time anyone uses it, not just its author

Notes on SkillClaw: Let Skills Evolve Collectively with Agentic Evolver (arXiv:2604.08377) — Ziyu Ma, Shidong Yang, Yuxiang Ji, Xucong Wang, Yong Wang, Yiming Hu, Tongwen Huang, Xiangxiang Chu · April 2026

Note published July 23, 2026 · written by SkillFed’s research pipeline from the paper above · how these notes are made

Most agent skills ship once and sit still. The same tool-usage mistake, the same brittle workaround, the same failure mode gets rediscovered independently by every user who hits it, because nothing carries that experience back into the skill itself. SkillClaw closes the loop by treating every session as signal: it pools the trajectories — what an agent tried and how it went — across every user of a skill, then runs an autonomous evolver over that pooled history. The evolver hunts for behavioral patterns that recur across different people's sessions, then decides whether the fix is tightening an existing skill or adding a new one. Updated skills land in a shared skill repository that syncs to every user automatically — nobody has to notice a fix shipped, let alone go pull it.

WildClawBench, a benchmark of real-world agent scenarios built for exactly this dynamic, provides the test bed, with Qwen3-Max running as the backbone model. Performance improves significantly — and each individual user contributes only a small amount of interaction and feedback to get there. That's the mechanism doing the work, not any one account's data volume: the system compounds thin per-user signal by pooling it across the whole population, instead of waiting for a single user to repeat the same mistake enough times to force a fix on their own.