DisCo distills 1,000 ML repos into 5,000+ skills, boosting MLE-bench by 134.3%
TheTuringPost · x · 2026-09-05
The Repo-To-Skill paper introduces DisCo, a skill-powered research agent that distills operational knowledge from GitHub repos into reusable, verified skill packages. Its AREX-Skill library packs 5,000+ skills from 1,000 popular ML repos across 20 areas and 178 capability families. With a fixed GPT-5.5 backbone and execution budget, skill-equipped agents score 134.3% higher on MLE-bench and 34.4% higher on PaperBench.
Related event: DisCo Distills 5000+ Skills from ML Repos, Boosting Agent Benchmark Scores(2 posts)→
More from coding & agent
- GPT-6 Astra Auto-Preps Entire Final Cut Project: Import, Color Grade, Sync Clips — minchoi · 2026-09-05
- GPT-6 Astra Ultra Generates an App in 26 Minutes Using 10.46M Tokens — minchoi · 2026-09-05
- Devs call to abolish context compaction: agent engineering debate heats up — sull · 2026-09-05
- GPT-6 Astra arrives in GitHub Copilot, built for long-horizon autonomous coding — OpenAIDevs · 2026-09-05
- Hierarchical RAG plus persistent scratchpad beats FIFO windows for long-horizon agents — IgorCarron · 2026-09-05
- Astra-built site lays out GPT-6's pitch: research, modeling, coding, cross-app agents — reach_vb · 2026-09-05