Prompt2Skill Builds LLM Skills From a Single Prompt, +10.8 Average Across Four Domains
Bo Ni · hf · 2026-10-02
Problem
Skills are external artifacts LLMs consume at inference time, but expert-authored skills are costly, not optimized for specific model versions, and new tasks often lack suitable skills — while existing auto-optimization methods require curated in-distribution training sets.
Method
Prompt2Skill builds skills from natural-language task descriptions alone: it derives a task specification, discovers or synthesizes datasets, and refines the skill in a closed loop of reflective editing.
Results
Across question answering, reading comprehension, spreadsheet manipulation, and mathematical reasoning, Prompt2Skill consistently beats direct prompting on open-source and frontier models, averaging +10.8.
More from coding & agent
- Dev says Claude and Claude Code are getting worse day by day — Pavan_Belagatti · 2026-10-02
- Dev hooks Reels into Claude Code so he can doomscroll while the agent codes — yoimnotkesku · 2026-10-02
- Book a flight in 2 min or endure a 10-min AI agent call? Devs argue agent UX — jxnlco · 2026-10-02
- How llama.cpp Finds Safe Seams: Inside the Model-Loading Compatibility Hub — Mahmoud_Zalt · 2026-10-02
- Argo-Bench Pits Data Agents Against a 7.5B-Row Warehouse; Best Model Clears Only 34.8% of Tasks — textql · 2026-10-02
- Flexport launches MCP letting AI tools ship containers worldwide, customs included — ycombinator · 2026-10-02