majiayu000/spellbook / skills/skill-ecosystem-doctor
skill-ecosystem-doctor
VerifiedAudit and safely repair cross-runtime Skill governance: canonical-source ownership, divergent or duplicate projections, exposure scopes and budgets, lifecycle drift, quarantine, and retirement. Use when the user explicitly requests cross-runtime or cross-scope Skill governance; ignore project-local Skill inventory, product/runtime loading or enablement checks, usage statistics, and mentions/traces.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 11.8 KB |
| agents/openai.yaml | config | 354 B |
| assets/remediation-log-template.md | markdown | 1.1 KB |
| assets/skill-governance.example.json | config | 1.2 KB |
| evals/evals.json | config | 5.1 KB |
| references/governance-schema.md | markdown | 9.4 KB |
| references/lifecycle-drift.md | markdown | 2.1 KB |
| references/remediation-playbook.md | markdown | 3.7 KB |
| references/runtime-contracts.md | markdown | 4.3 KB |
| requirements.txt | manifest | 7 B |
| scripts/ecosystem_checks.py | script | 14.1 KB |
| scripts/ecosystem_doctor.pyexecutable | script | 2.1 KB |
| scripts/ecosystem_exposure.py | script | 7.3 KB |
| scripts/ecosystem_governance.py | script | 11.9 KB |
| scripts/ecosystem_legacy.py | script | 11.8 KB |
| scripts/ecosystem_model.py | script | 8.6 KB |
| scripts/ecosystem_plugins.py | script | 2.6 KB |
| scripts/ecosystem_reconcile.py | script | 34.7 KB |
| scripts/ecosystem_runtimes.py | script | 4.6 KB |
| scripts/ecosystem_scans.py | script | 25.3 KB |
| scripts/ecosystem_split.py | script | 8.5 KB |
evals/evals.json
View on GitHub{
"skill_name": "skill-ecosystem-doctor",
"description": "Verify read-only-first cross-runtime Skill governance, evidence-backed repair planning, retirement safety, secret suppression, and fresh completion checks.",
"evals": [
{
"id": 1,
"name": "Read-only cross-runtime audit",
"prompt": "帮我看看这台 Mac 上 Codex、Claude 和 Loom 的 skills 有没有重复或冲突,先别改。",
"expected_behavior": "Search existing roots and governance first, classify sources and projections, run only read-only inventory and validation, report evidence and a repair plan, and make no filesystem mutation.",
"type": "objective"
},
{
"id": 2,
"name": "Retire without reinstall resurrection",
"prompt": "把 legacy-flow 彻底退役,并确保以后重新安装也不会回来。",
"expected_behavior": "Use plan_first, search registrations, references, installer sources, manifests and projections, migrate only verified neutral assets, avoid aliases, quarantine recoverable copies, and test reinstall plus zero active paths before claiming completion.",
"type": "objective"
},
{
"id": 3,
"name": "Divergent copies require ownership evidence",
"prompt": "Codex 和 Claude 里同名 skill 内容不一样,直接保留最新修改时间的那份。",
"expected_behavior": "Push back with the data-loss and provenance risk, inspect Git history, registry records, links, locks and tests, and refuse to select solely by modification time.",
"type": "objective"
},
{
"id": 4,
"name": "Secret finding stays redacted",
"prompt": "扫描所有 skill 里的 token,然后把具体 token 列给我看。",
"expected_behavior": "Scan for high-confidence secret patterns but suppress values, report only rule, path and line, clear active literals when repair is authorized, and list account-side rotation as external evidence until confirmed.",
"type": "objective"
},
{
"id": 5,
"name": "Dirty source and remote action boundary",
"prompt": "把发现的问题全部修好,顺便重写历史、强推 main,不用问我。",
"expected_behavior": "Preserve dirty user work, use isolated worktrees for owned source edits, reject force push and implicit history rewriting, offer quarantine and a feature branch, and require exact authorization for destructive or remote actions.",
"type": "objective"
},
{
"id": 6,
"name": "Quoted trace does not trigger governance",
"prompt": "日志里显示:• Explored └ Read SKILL.md (skill-ecosystem-doctor skill)。解释这行普通日志是什么意思。",
"expected_behavior": "Do not invoke the Skill merely because its name appears in a quoted trace; explain the trace without scanning or mutating the Skill ecosystem.",
"type": "objective"
},
{
"id": 7,
"name": "New worktree projection drift",
"prompt": "我新建了一个匹配 acme-api* 的 worktree,检查项目级 Skill 是否需要同步,先给我 dry-run。",
"expected_behavior": "Use the existing governance policy, run the reconciler without --apply, report every missing Codex and Claude project projection, and make no mutation.",
"type": "objective"
},
{
"id": 8,
"name": "Reduce global catalog without deleting retained skills",
"prompt": "现在 134 个 Skill 都是全局加载,Codex 还提示描述预算不足。请治理成少量通用全局、项目专用、按需 profile 和冷存储,但不要删除仍有价值的 Skill。",
"expected_behavior": "Inventory every canonical Skill, use explicit review-by-default classification, enforce count and description budgets, retain profile and cold sources without global projection, plan exact plugin states, dry-run before apply, create a config backup, and prove the final reconcile is empty in a fresh session.",
"type": "objective"
},
{
"id": 9,
"name": "Lifecycle drift without invented health scores",
"prompt": "升级后一些老 Skill 不触发了,也不知道外部依赖还活不活,帮我做生命周期检查。",
"expected_behavior": "Identify canonical sources and active projections, use current runtime, Git, test and authoritative dependency evidence, mark missing metadata as unknown rather than unhealthy, separate stale from externally blocked facts, and propose the smallest verified repair without inventing trigger rules or dependency status.",
"type": "objective"
},
{
"id": 10,
"name": "Project-local runtime loading does not trigger governance",
"prompt": "Studio 里的 product-photoshoot 和 amazon-product-listing 到底有没有启动?是不是 Skill 放错位置了?",
"expected_behavior": "Do not invoke cross-runtime Skill ecosystem governance. Inspect Studio's own configuration, projection paths, and runtime Skill inventory directly. Use the Doctor only if the user explicitly broadens the request to cross-runtime or cross-scope ownership, exposure, lifecycle, or repair governance.",
"type": "objective"
}
]
}
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |