SkillOpt: Self-Evolving Agent Skills with Zero Inference Overhead Microsoft released SkillOpt v0.2.0 on PyPI on 2026-07-02, adding SkillOpt-Sleep, a nightly offline self-evolution engine that runs harvest, mine, replay and consolidate stages behind a held-out validation gate and ships as the skillopt-sleep CLI. The release also adds integration shells for Claude Code, Codex, Copilot and Devin plus an OpenClaw reference adaptation, which live in the repository rather than the PyPI wheel, along with SearchQA split materialization, Windows robustness fixes and hardened JSON parsing. SkillOpt trains agent skills like neural networks β€” with epochs, batch sizes, learning rates and validation gates β€” without touching model weights, and was accepted to NeurIPS 2026 as a Spotlight and won 1st Place in the Executive Challenge at the Microsoft Global Hackathon 2026. Train agent skills like you train neural networks β€” with epochs, mini- batchsize, learning rates, and validation gates β€” but without touching model weights. πŸ“– For installation, data preparation, training/eval commands, configuration, and framework internals, start with the versioned SkillOpt documentation https://github.com/microsoft/SkillOpt/blob/main/docs/index.md . A concise rendered overview is available in the Documentation & Reproduction Guide https://microsoft.github.io/SkillOpt/docs/guideline.html , and longer-form engineering analysis appears on the Technical Blog https://microsoft.github.io/SkillOpt/blog/ . We also maintain a Changelog https://github.com/microsoft/SkillOpt/blob/main/CHANGELOG.md for released and unreleased changes. - 2026-11 🎀 SkillOpt is accepted to NeurIPS 2026 as a Spotlight See you in Sydney - 2026-11 πŸ† SkillOpt won 1st Place in the Executive Challenge at the Microsoft Global Hackathon 2026 badge https://www.credly.com/badges/d9fd6cd8-e310-4e9b-9f63-ed7a3383fc85/public url , issued by The Microsoft Garage . - 2026-07-24 πŸ“° SkillOpt in the news. Read the official Microsoft Research feature https://www.microsoft.com/en-us/research/blog/skillopt-agent-skills-as-trainable-parameters/ , along with recent coverage from VentureBeat https://venturebeat.com/orchestration/microsofts-open-source-skillopt-automatically-upgrades-ai-agent-skills-without-touching-model-weights , Synced ζœΊε™¨δΉ‹εΏƒ https://mp.weixin.qq.com/s/pMlyj3a3KOh8L7cIHClRXA , Flowtivity https://flowtivity.ai/blog/microsoft-skillopt-train-ai-agent-skills/ , and The Decoder https://the-decoder.com/microsofts-skillopt-boosts-gpt-5-5-by-using-nothing-but-a-trained-markdown-file/ . - 2026-07-02 πŸš€ SkillOpt v0.2.0 https://github.com/microsoft/SkillOpt/releases/tag/v0.2.0 is out on PyPI https://pypi.org/project/skillopt/ Headline feature: SkillOpt-Sleep , a nightly offline self-evolution engine harvest β†’ mine β†’ replay β†’ consolidate behind a held-out validation gate , now shipped as the skillopt-sleep CLI. It also includes experimental multi-objective, replay, and dream-rollout controls; the main CLI keeps conservative defaults and does not expose every experiment-harness control as a flag. The release source adds integration shells for Claude Code, Codex, Copilot, and Devin , plus an OpenClaw reference adaptation ; these plugin/MCP files live in the repository rather than the PyPI wheel. It also adds SearchQA split materialization, Windows robustness, and hardened JSON parsing. See the release notes https://github.com/microsoft/SkillOpt/releases/tag/v0.2.0 for full release details and contributor acknowledgements. - 2026-06-15 😴 SkillOpt-Sleep preview β€” a nightly offline self-evolution companion for local coding agents Claude Code / Codex / Copilot : review past sessions, replay recurring tasks, and consolidate validated skills behind a held-out gate. See docs/sleep/README.md https://github.com/microsoft/SkillOpt/blob/main/docs/sleep/README.md for what it is, how to use it, and results. - 2026-06-03 πŸŽ‰ gbrain https://github.com/garrytan/gbrain , gbrain-evals https://github.com/garrytan/gbrain-evals/blob/main/docs/benchmarks/2026-06-03-skillopt.md , and darwin-skill https://github.com/alchaincyf/darwin-skill have all integrated SkillOpt. - 2026-06-02 πŸŽ‰ SkillOpt v0.1.0 https://github.com/microsoft/SkillOpt/releases/tag/v0.1.0 is now available on PyPI https://pypi.org/project/skillopt/ Install with pip install skillopt . This initial release includes the full training loop rollout β†’ reflect β†’ aggregate β†’ select β†’ update β†’ evaluate , multi-backend support OpenAI / Azure / Claude / Qwen / MiniMax , six built-in benchmarks, and WebUI dashboard. Modern agent skills are usually hand-crafted, generated one-shot by a strong LLM, or evolved through loosely controlled self-revision β€” none of which behaves like a deep-learning optimizer for the skill itself, and none of which reliably improves over its starting point under feedback. SkillOpt treats the skill document as the trainable state of a frozen agent , and trains it with the discipline that makes weight-space optimization reproducible. A separate optimizer model turns scored rollouts into bounded add / delete / replace edits on a single skill document; in the default paper-style path, a candidate edit is accepted only when it strictly improves a held-out validation score. A textual learning-rate budget, a rejected-edit buffer, and an epoch-wise slow / meta update make skill training stable while adding zero inference-time model calls at deployment. The deployed artifact is a compact best skill.md typically 300–2,000 tokens that runs against the unchanged target model. Across six benchmarks, seven target models, and three execution harnesses direct chat, Codex CLI, Claude Code CLI , SkillOpt is best or tied-best on all 52 evaluated model, benchmark, harness cells and on GPT-5.5 lifts the average no-skill accuracy by +23.5 points in direct chat, +24.8 inside the Codex agentic loop, and +19.1 inside Claude Code . Optimized skill artifacts transfer across model scales, between Codex and Claude Code harnesses, and to nearby benchmarks without further optimization. For the full method, ablations, and per-cell results see the paper https://arxiv.org/abs/2605.23904 ; for a visual walkthrough of the loop see the project page https://microsoft.github.io/SkillOpt/ ; for deeper API / backend / benchmark docs see docs/ https://github.com/microsoft/SkillOpt/blob/main/docs . 64c8f76086bed7bd7a5ce664a7a14f40 raw.mp4 β–Ά Watch the full demo on YouTube A backend = a chat / exec target e.g. openai chat , claude chat , qwen chat , minimax chat , copilot chat , openai compatible , codex exec , claude code exec , cursor exec , copilot exec . If a provider implements the OpenAI Chat Completions protocol, try the built-in openai compatible backend before adding code. See docs/guide/new-backend.md https://github.com/microsoft/SkillOpt/blob/main/docs/guide/new-backend.md for the full contract. Chat backends add a skillopt/model/