skillopt
Skill1 starsChanged 10 days agoMITView on GitHub
Runs Microsoft SkillOpt training/optimization for agent skill prompts inside the isolated .agents/venv_skillopt/ environment. Every optimizer run requires explicit human authorization.
- Installs packages
5 other projects use a version of this.
Adopted by
Most recently taken up by
- mannyandrs45/agentic-skill-librarytheir own version, first seen Oct 15, 2026
- iliaal/whetstonetheir own version, first seen Jul 12, 2026
- skillberry-ai/cap-evolvetheir own version, first seen Jun 16, 2026
- TheophilusChinomona/idevtheir own version, first seen Jun 12, 2026
- harrylabsj/skillopttheir own version, first seen Jun 7, 2026
Other versions5 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 5
- skillberry-ai/cap-evolve★ 69same name
- iliaal/whetstone★ 36same name
- harrylabsj/skillopt★ 2same name
- mannyandrs45/agentic-skill-library★ 0same name
- TheophilusChinomona/idev★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository37 other files it gives its agents.
- agents.mdAGENTS.md
- skills/vercel-composition-patterns/AGENTS.mdAGENTS.md
- skills/vercel-react-best-practices/AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- accessibilitySkill
- autoskillsSkill
- compliance-checkerSkill
- contract-writerSkill
- django-expertSkill
- django-patternsSkill
- django-securitySkill
- django-tddSkill
- django-verificationSkill
- env-shielding-auditorSkill
- frontend-designSkill
- graphifySkill
- js-standardizerSkill
- mass-standardizerSkill
- mcp-registrySkill
- nodejs-backend-patternsSkill
- nodejs-best-practicesSkill
- omni-context-minimizerSkill
- python-quality-auditorSkill
- readme-standardizerSkill
- seoSkill
- skill-creatorSkill
- skills-samplesSkill
- slash-commanderSkill
- sprint-architectSkill
- tailwind-css-patternsSkill
- token-saver-auditorSkill
- topology-monitorSkill
- topology-scaffolderSkill
- typescript-advanced-typesSkill
- vercel-composition-patternsSkill
- vercel-react-best-practicesSkill
- viteSkill
The file
We split this skill into the rules it gives an agent
1 rule. 0 are common, 1 is distinctive, and 0 are very specific to this project.
0
Common
Most of the 4 edited versions we read keep these.
1
Distinctive
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 32
Explicit Authorization: Every execution of the optimizer train loop must be approved by the human operator.
“Common” is measured: kept by at least 6 in 10 of the 4 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
🛠️ Skill: SkillOpt (Rule 71 Compliance)
Domain
- Category: Infrastructure / Optimization
- Origin: Third-Party Concept Integration (Microsoft Research)
- Status:
ACTIVE_LOCAL
Technical Logic
This tool integrates Microsoft's SkillOpt using a separate isolated virtual environment (.agents/venv_skillopt/) and runtime monkeypatching. It registers the agents_opt benchmark adapter and routes model generation to gemini_backend.py without mutating PyPI package binaries.
Provisioning (on demand)
The session bootstrap only installs the lean core (requirements-core.txt). Before the FIRST training run in an environment, install this skill's heavy stack:
.agents/venv_skillopt/bin/pip install -r .agents/requirements-skillopt.txtThis transitively pulls in azure-* packages (recorded in docs/audits/SKILLOPT_TRANSITIVE_CLOSURE.md, which replaced requirements-freeze.txt in Sprint 023 C7 so that a scanner stops reading an optional closure as this framework's own manifest) — the vendored skillopt package ships an azure_openai backend module by default. It is never called: train_runner.py apply_monkeypatches() imports that module only as a patch target and overwrites its functions to route to gemini_backend.py (or the Claude Agent SDK, via model_backend: claude in the config — see configs/agents_opt.yaml). Do not "clean up" the azure packages; they are a real, if unused, transitive dependency of the pinned skillopt version.
Commands
# Execute optimization using configs
.agents/venv_skillopt/bin/python .agents/skills/skillopt/scripts/train_runner.py --config .agents/skills/skillopt/configs/agents_opt.yamlGovernance Audit
- Rule 71 Compliance (Flat Mapping): Replaced nested folder clones with PyPI dependencies and flat scripts under
.agents/skills/skillopt/. - Double Venv Isolation: Mandated to run commands exclusively within
.agents/venv_skillopt/to protect Django dependencies. - Explicit Authorization: Every execution of the optimizer train loop must be approved by the human operator.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.