Skip to content
AI HOT (Curated Pool)

Skill distillation

技能提炼

Skill distillation has Opus 4.7, GPT-5.1, and Gemini 3 Pro write standardized SKILL.md procedure files, while local Qwen 35B and Gemma 26B models execute those files step by step.

Why it matters: HKR-H/K/R pass: the agent-skill distillation pattern is concrete and practitioner-relevant. The summary lacks success rates, cost data, or task outcomes, so it sits at the featured threshold, not must-write.

Read the original ↗Export Markdown