Distilled SKILL.md beats Workflow Memory by 6.06 points — skills stabilize execution, not knowledge
rohanpaul_ai · x · 2026-09-08
In matched experiments, agents given the same past trajectories as a distilled SKILL.md outperformed Workflow Memory (which keeps more execution detail) by 6.06 percentage points — the gain came from better packaging of identical experience, not more experience.
Trajectory analysis clarifies the mechanism: 65.7% of successful skill cases worked through "procedural anchoring" — stabilizing what to do first, which tools to use, and what to verify — while only 4.5% worked by injecting missing knowledge. Skills mainly fix execution, not knowledge gaps.
This also explains the failure mode: skills hurt when applied in the wrong context or followed too rigidly. The takeaway: self-improving agents need better distillation and application of experience, not bigger memory libraries.
Related event: 8,135 Experiments Demystify When Agent Skills Work and Fail(2 posts)→
More from coding & agent
- AI code review blasted for flagging Carmack's inverse sqrt in 10 lines of working code — ZeeshanZiaML · 2026-09-08
- DHH's Omarchy Linux launches foundation with $15.5M, betting on agent-powered troubleshooting — vista8 · 2026-09-08
- CROCODIL tackles over-editing when LLMs modify code written by other models — jessyjli · 2026-09-08
- How to Auto-Generate Figma Photomosaics with Grok Bot, Cutting 90% of the Grunt Work — mattyp · 2026-09-08
- x402 gives agents price tags per call, enabling economic reasoning, says Coinbase dev lead — kleffew94 · 2026-09-08
- Writer uses qwen3.8:27b to audit his novella and builds a git hook blocking non-ascii commits — walkingriver · 2026-09-08