Yao meta skill
Create/improve/evaluate agent skills from workflows, prompts, SOPs, scripts. Use for migration/release/package, routing, evals/tests, install checks, 优化已有 skill, 补 trigger 评测. Exclude no-skill summary/translation/docs.From its SKILL.md
npx -y skills add yaojingang/yao-meta-skillAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- runs commandsInstructs the agent to run 1 command, including `trigger_eval.py`.
SKILL.md
3.1 KB, 666 tokens by cl100k_base, as published. Nobody here has run it
Yao Meta Skill
Router Rules
- Route by frontmatter
description. - Keep
SKILL.mdlean; put guidance inreferences/, logic inscripts/, and evidence inreports/. - Use the lightest reliable process.
Modes
Scaffold: exploratory/personal.Production: team reuse.Library: shared infra.Governed: high-trust/release-critical.- Rules: Method, Operating Modes, Resource Boundaries.
Compact Workflow
- One-off/no reusable process:
Do not create a skill;near-neighbor; requirerepeated use+reusable output contract. - Capture job, output, exclusions, constraints, standards, lightest fit.
- Scan up to
3-5references: external, user, local; skip tiny edits; surface only uncertainty/conflict. - Write
descriptionearly; route/boundary edits needtrigger_eval.py; releases need risk-matched gates before folders. - Add output-risk, artifact-design, prompt-quality, system-model, next directions only when earned.
Playbooks: Method, Intent, Skill IR, Output Eval, Review Studio.
Skill OS 2.0 Gates
For production/library/governed/team releases, run Skill IR, compiler, trigger/output eval, Skill Atlas, conformance, trust, registry/package/install, upgrade, drift, waiver, Review Studio.
Governed Package Boundary
For file-backed/release-critical/governed packages, name input_files as file-backed fixture; include owner, review cadence, input_files, output contract, rollback boundary; require trust report and reports/output_quality_scorecard.md; mark unavailable telemetry/approvals/metrics/benchmarks as missing evidence; do not fabricate evidence.
Preserve labels literally when they apply: file-backed fixture, input_files, output contract, rollback boundary, trust report, reports/output_quality_scorecard.md, missing evidence.
First-Turn Style
- Start from the user's work/outcome before structure.
- Ask only
2-3key questions unless enough detail exists. - In Chinese, sound soft and companion-like; use Intent Dialogue.
Output Contract
Create/refactor/package: produce SKILL.md, aligned agents/interface.yaml, justified assets, boundary/exclusion/gate summary. Audit/evaluate-only: findings + proposed fixes; edit files only if asked. No-skill: no files.
Reference Map
Primary: Method, Artifact Design, Systems, Governance, SkillOps.
What ships with it: 652 files
11534.6 KB alongside SKILL.md, 236 of them executable
agents/
- interface.yaml774 B
assets/
- review-studio.css20.2 KB
- review-viewer.css3.1 KB
- skill-overview.css18.5 KB
- skill-overview.jsruns1.8 KB
docs/
- assets/social-preview.png47.1 KB
- assets/social-preview.svg3.6 KB
- migration-v2.md2.7 KB
- README.fr-FR.md14.5 KB
- README.ja-JP.md14.7 KB
- README.ru-RU.md19.0 KB
- README.zh-CN.md21.3 KB
evals/
- adaptation/user_signals.example.jsonl743 B
- adversarial/trigger_cases.json1.2 KB
- baseline_description.txt33 B
- blind_holdout/trigger_cases.json1.2 KB
- confusion/route_cases.json3.2 KB
- dev/trigger_cases.json3.8 KB
- failure-cases.md889 B
- history/2026-03-31-context-first-pack.json686 B
- history/2026-03-31-family-suite-expansion.json635 B
- history/2026-03-31-governance-quality-gates.json654 B
- history/2026-03-31-governed-assets-and-history.json699 B
- history/2026-04-01-route-scorecard-foundation.json769 B
- history/description_optimization/2026-03-31-description-optimization-suite.json1.4 KB
- history/description_optimization/2026-04-01-adversarial-calibration-and-family-drift.json11.8 KB
- history/description_optimization/2026-04-01-blind-holdout-and-drift-history.json3.0 KB
- history/description_optimization/2026-04-01-judge-backed-blind-eval.json14.8 KB
- history/description_optimization/README.md660 B
- history/README.md440 B
- holdout/trigger_cases.json2.4 KB
- improved_description.txt233 B
- output/cases.jsonl6.4 KB
- output/fixtures/release-brief-source.md387 B
- output/schema.json2.1 KB
- packaging_expectations.json1.0 KB
- promotion_policy.md2.1 KB
- README.md6.6 KB
- semantic_config.json11.6 KB
- AGENTS.md14.2 KB
612 more files not listed here. See all 652 in the repository.