Skill∞: Recursive Self-Improvement Where Agents Optimize Their Own Optimization Skills
NationalUniversityofSingapore · hf · 2026-10-27
A paper from NUS and City University of Hong Kong, Skill∞ tackles recursive self-improvement: existing skill optimizers follow hand-designed procedures and lock meta-skill evolution into fixed pipelines. Skill∞ encodes both the improvement target and the improving procedure as skills, letting a meta agent recursively optimize task skills and its own meta-skill through a reusable, agent-led process across domains, harness-agnostic across Claude Code, Codex, and more.
More from coding & agent
- shadcn: a voice dictation app is just T3 code plus one prompt — shadcn · 2026-10-27
- Coding got cheap, but the friction we removed was doing useful cognitive work — charles_irl · 2026-10-27
- One dev + AI builds open-source Rust suite targeting Photoshop, Premiere and Office — ankur_maker · 2026-10-27
- Anthropic: Single agents with skills often beat subagent designs in enterprise deployments — addyosmani · 2026-10-27
- Claude Code's idle compaction annoys users — here's the setting to turn it off — JasonBotterill · 2026-10-27
- Creator builds CCTV-style video intro with Claude Code, still stuck on AI dubbing — xiaohu · 2026-10-27