GH Repository · microsoft
SkillOpt
SkillOpt is a text-space optimizer that trains reusable natural-language skills for frozen LLM agents through trajectory-driven edits, validation-gated updates, and deployable best_skill.md artifacts.
- stars
- 17,099
- 30-day movement
- +18863/day
- Related entries
- 13
- Connections
- 1
pythonskill-packPythonself-evolving-agentsagent-skills
SkillOpt is a text-space optimizer from Microsoft that trains reusable natural-language skills for frozen LLM agents. It edits skills based on agent trajectories, gates updates through validation, and produces a deployable best_skill.md artifact.
Use it to improve a frozen LLM agent's behavior by optimizing prompt-level skills instead of model weights.
Use it to
- Train reusable skills from agent trajectories
- Gate skill updates via validation
- Deploy best_skill.md to your agent
- Optimize prompts without touching model weights
For Developers building and tuning LLM agent systems
- Role
- skill-pack
- Language
- Python
- Licence
- MIT
- Forks
- 1,598
- Open issues
- 20
- Last push
- 2026-09-05
- Latest release
- v0.1.0 · 2026-06-02
topicsagent-skillsself-evolving-agentstext-space-optimizationprompt-optimizationllm-agentspython