Skip to content

Adonis0123/adonis-skills / skills/architecture-hardening-loop

architecture-hardening-loop

Verified

Run a bounded scan-triage-fix-independent-review-rescan loop on an explicit or uniquely resolvable user-supplied code scope until no evidence-backed architecture fixes remain. Use for implementation-inclusive architecture cleanup, DDD or high-cohesion hardening, and autonomous architecture improvement.

FileKindSize
SKILL.mdskill md25.1 KB
agents/openai.yamlconfig713 B
evals/evals.jsonconfig40.3 KB
evals/trigger-eval.jsonconfig4.4 KB
references/ownership-and-evidence.mdmarkdown6.7 KB
references/scope-resolution.mdmarkdown1.6 KB

evals/trigger-eval.json

View on GitHub
[
  {
    "query": "把范围冻在 packages/video-editor/core/**,跑一轮 architecture scan、证据分类、最小修复、测试、Grok review 和同范围 rescan,直到 NO_ACTIONABLE_FINDINGS。",
    "should_trigger": true
  },
  {
    "query": "Use architecture-hardening-loop on apps/web/src/features/billing/**. Fix only evidence-backed module-boundary issues, review every batch, and rescan the same scope until clean.",
    "should_trigger": true
  },
  {
    "query": "scanner 已经给出 packages/core 的候选报告;现在按 Fix/Backlog/Reject 过证据门槛,把值得修的做掉,验完让 Grok 复审,再扫原目录。",
    "should_trigger": true
  },
  {
    "query": "对 src/order/** 做高内聚低耦合加固,不要只给建议:有当前伤害的才最小修,拒绝未来式抽象,并循环 review/rescan 到没有 actionable finding。",
    "should_trigger": true
  },
  {
    "query": "我们把 scope 限定在 packages/payments/providers/**。请自主做 DDD boundary hardening、回归测试和独立 review,最多三轮,不要碰 checkout UI。",
    "should_trigger": true
  },
  {
    "query": "clean up the architecture in src/editor/timeline/** in a bounded improve-review-fix cycle; preserve public behavior and stop only after a same-scope rescan finds no Fix.",
    "should_trigger": true
  },
  {
    "query": "只看 services/catalog 与 services/search 的依赖方向,把可复现的循环依赖修掉,每批跑测试、Grok review,再重新扫描这两个模块。",
    "should_trigger": true
  },
  {
    "query": "对 packages/shared-auth/** 做 scan-fix-review-until-clean;明确 Reject 纯六边形洁癖,任何 scope 扩张都停下来问我。",
    "should_trigger": true
  },
  {
    "query": "I want an implementation-inclusive architecture review of lib/persistence/**: evidence, impact, minimal change, verification, peer review, then rescan until no actionable findings remain.",
    "should_trigger": true
  },
  {
    "query": "在 modules/notifications/** 自主加固架构,修复真实传播成本问题并补测试;每轮保持同一 scope,最终报告 Backlog/Reject 和 NO_ACTIONABLE_FINDINGS。",
    "should_trigger": true
  },
  {
    "query": "只读分析 packages/video-editor/core/** 的模块边界,生成一份 architecture report;不要修改、不要进入 fix loop。",
    "should_trigger": false
  },
  {
    "query": "在 apps/web 增加一个已定稿的 Share modal,按 docs/share-modal-spec.md 直接实现和测试,不需要架构扫描。",
    "should_trigger": false
  },
  {
    "query": "review 一下当前 staged diff,发现问题就修,继续 re-review 到 PASS;不需要 architecture scanner 或同范围 rescan。",
    "should_trigger": false
  },
  {
    "query": "请 fresh eyes 看 feature/billing-redesign 的 diff,只做一次只读 code review,把 findings 按严重度列出来。",
    "should_trigger": false
  },
  {
    "query": "对 docs/plans/project-persistence.md 走完整三 agent completion:实现所有 PENDING、Goal、review、architecture gate 和 fresh Claude audit。",
    "should_trigger": false
  },
  {
    "query": "Use architecture-hardening-loop to clean this repository, but I haven't chosen any path, package, module, or exclusions yet.",
    "should_trigger": false
  },
  {
    "query": "先讨论订单域应该怎么拆 aggregate、Repository 接口是否需要;现在没有实现 scope,也不要改代码。",
    "should_trigger": false
  },
  {
    "query": "packages/core 的测试在 CI 偶发超时但本地正常,先定位根因并修复,不要做架构清理。",
    "should_trigger": false
  },
  {
    "query": "把 src/order/cancel.ts 里的重复校验提成一个 helper,行为和签名都已经给定,只改这个文件并跑单测。",
    "should_trigger": false
  },
  {
    "query": "profile 一下 editor playback 为什么掉帧,先给 flamegraph 和根因,不要改模块边界或进入 scan-fix loop。",
    "should_trigger": false
  },
  {
    "query": "用 architecture-hardening-loop 看最近 3 个 commit 触及的文件,加上我还没提交的改动,直接修到没有可修项;不用再问我 hash。",
    "should_trigger": true
  },
  {
    "query": "attach 里这两个提交(标题我截图了)做一轮架构加固:只修有当前伤害的,Grok 复审后同范围复扫。",
    "should_trigger": true
  }
]
receiving-code-reviewobra/superpowersUse when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement or blind implementationStars: 322kUpdated Claude CodeCodexVerifiedInstall
requesting-code-reviewobra/superpowersUse when completing tasks, implementing major features, or before merging to verify work meets requirementsStars: 322kUpdated Claude CodeCodexVerifiedInstall
using-git-worktreesobra/superpowersUse when starting feature work that needs isolation from current workspace or before executing implementation plans - ensures an isolated workspace exists via native tools or git worktree fallbackStars: 322kUpdated Claude CodeCodexVerifiedInstall
verification-before-completionobra/superpowersUse when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions alwaysStars: 322kUpdated Claude CodeCodexVerifiedInstall