Microsoft Open Sources SkillOpt: Train Agent Skills Like Neural Networks
aigclink · x · 2026-08-31
Microsoft has open-sourced SkillOpt, a text-space optimizer that trains reusable natural-language skills for frozen LLM agents through trajectory-driven edits and validation-gated updates, all without touching model weights. It applies neural network training concepts (epochs, batch size, learning rates) to optimize skill documentation, outputting a compact bestskill.md.
Key Features:
- SkillOpt-Sleep: An offline self-evolution feature that harvests daily session data to mine repetitive tasks for replay practice, merging tested skills into the guide.
- Zero Inference Overhead: Adds no extra inference calls or latency during deployment.
- Performance: Achieved a 23.5-point improvement over the no-skill baseline in GPT-5.5 direct chat scenarios.
More from coding & agent
- Simon Willison: code mode is the future of tool calling, if it can be sandboxed — springrod · 2026-10-16
- Dev workflow: Astra for tedious tasks, Claude for ideas not implementation — menhguin · 2026-10-16
- Agent named 'Ilya' reportedly scores 1.1pt higher on benchmarks — daniel_mac8 · 2026-10-16
- Gemini CLI fix: substring matching inlined unrequested binary files into model context — priyanshu1976 · 2026-10-16
- Building AI agents on real healthcare claims data: the domain work that makes them work — alex_verem · 2026-10-16
- OpenAI stealth-testing mysterious internal model 'Aeon' in Codex desktop app and Dots — cedric_chee · 2026-10-16