All categories
Code Review & Quality
Code review, refactoring, debugging, maintainability, and engineering quality
- 15,936
- 664
Security grades appear on each card once the skill has been scanned. Newly imported skills may briefly show without a grade until the backfill job runs.
Open in full browserBrowse code review & quality skills
Showing 15,313–15,336 of 15,936 skills
- Pr Review RubricStaff-level PR review rubric — correctness, contract, clarity, tests, blast radius. Use when reviewing code (yours or others') before merge.Votes: 0GitHub stars: 2
- Debug PlaybookHypothesis-driven debugging — reproduce, hypothesize, probe, locate, fix, guard. Use when something is broken and you don't yet know why.Votes: 0GitHub stars: 2
- Wiki LintRun a deterministic, read-only health check on an Obsidian wiki. Use for lint, vault health check, audit wiki health, find orphans, find dead links, frontmatter audit, provenance audit, or wiki audit. Reports graph, link, frontmatter, provenance-ledger, empty-section, and stale-index findings; it does not reason broadly or repair files.Votes: 0GitHub stars: 15,078
- Review PrComprehensive code review using specialized agents. Each agent focuses on a different quality aspect (code, errors, comments, types, tests, simplify). It runs the paired-session report mode (a PR, a ref or the local change; no writer, nothing posted); the legacy review was removed in v2.13.0, so `entry: legacy` and `--legacy` are refused.Votes: 0GitHub stars: 3
- Review LoopCodex-native Stage 1 review-loop skill. Orchestrates planning and execution with a Codex Executor and Claude/Codex reviewer backends while sharing the .review-loop protocol with Claude Code.Votes: 0GitHub stars: 3
- PlanLegacy plan skill (legacy workflow (removed from routing in v2.13.0; deleted in v2.13.1)). Use only when the user explicitly names the legacy plan skill; a plan request otherwise goes to review-loop or paired-session. Codex Stage 1 planning-only skill. Drives a work item from raw description to a reviewer-approved plan in `.review-loop/sessions/{uuid}.md`, then exits with a hint to resume via `review-loop:execute --session UUID`. Use when you want plan-only iteration without immediately enter...Votes: 0GitHub stars: 3
- ExecuteLegacy execute skill (legacy workflow (removed from routing in v2.13.0; deleted in v2.13.1)). Use only when the user explicitly names the legacy execute skill; an existing plan or a review of existing code otherwise goes to review-loop. Codex Stage 1 execute skill. Run the execution + quality polish + delivery stages of review-loop against one of three entry modes: resume an approved session (`--session`), execute a user-supplied plan (`--plan`), or run a pure-CR pass on the current working t...Votes: 0GitHub stars: 3
- ReviewReview spec-forge generated documents (SRS + tech-design + feature specs) for quality, completeness, and internal consistency. Finds issues like incomplete sections, contradictions, missing traceability, and vague specs, then optionally auto-fixes them. Supports iterative review-fix cycles that run until the review converges — passing, or a round with no net reduction in findings — typically 1–2 rounds.Votes: 0GitHub stars: 9
- Debug SkillA skill with debug mode enabledVotes: 0GitHub stars: 14
- GovernanceEnforces engineering standards and code quality policies. Use during code reviews, before commits, when discussing standards or compliance, for quality audits, and when running retrospectives. Trigger phrases include 'run a retro', 'retrospective', 'code review', 'run review', or 'audit standards'.Votes: 0GitHub stars: 10
- Openskills Runtime DebugDiagnose openskills-runtime execution failures in sandboxed paths (Landlock, seatbelt, native script execution, wasm execution) and produce root-cause-first findings with minimal-risk remediation steps.Votes: 0GitHub stars: 71
- AuditcodexdirectSend specific files to OpenAI Codex CLI for an independent audit/reviewVotes: 0GitHub stars: 201
- AuditcodexSend recent work to OpenAI Codex CLI for an independent audit/reviewVotes: 0GitHub stars: 201
- Mx ReviewReview code for MX Space project conventions. Checks NestJS patterns, Drizzle ORM repositories, Zod schemas, API design, etc.Votes: 0GitHub stars: 564
- Code Style程式碼風格規範。定義命名慣例、格式化規則、import 排序等。 所有開發類 agent 應注入此 skill。Votes: 0GitHub stars: 34
- AttentionAudit the Golden Path Clarity score for frontend projects. Evaluates whether each UI funnels users into a single monetizable journey or leaks attention across competing paths.Votes: 0GitHub stars: 7
- 02 Review子 skill — 代码评审与理解。当用户要 PR 评审、PR 写作（作者侧）、或要理解陌生代码模块的请求路径/调用栈时使用。包含 PR 评审（diff.html）、PR 写作（pr.html）、模块图三种范式。当父 skill 路由命中"评审/PR/审阅/这段代码怎么工作"类请求时加载。Votes: 0GitHub stars: 5
- Skill AuditSKILL.md SSL (Scheduling-Structural-Logical) auditor. Decomposes a target SKILL.md (or a directory of skills) into the 3-layer frame, surfaces missing / ambiguous / risky declarations, and emits a diagnostic report. Read-only — never modifies originals. Use for self-review before PR, refactor regression checks, or bulk repo audits. Triggers: "skill-audit", "skill audit", "ssl audit", "audit skill", "ssl 분해", "skill auditor", "audit-skill"Votes: 0GitHub stars: 3
- Self AuditMeta-gate that audits the current session's own actions before any commit — philosophy, scope, convention, and token-budget violations. Read-only on session state; emits a single inline table verdict. Use as the final step of any multi-edit / multi-cycle task. Triggers: "self-audit", "audit yourself", "self-check", "셀프 점검", "셀프 감사", "내 행동 점검", "before commit"Votes: 0GitHub stars: 3
- RalphiQuality-loop discipline. Enforces a per-cycle token budget (setup → worker → verifier → commit → handoff phases) and runs a between-cycle philosophy auto-check via rg sweep over the diff. Halts on any introduced ROP / version-hardcoding violation. Etymology — "ralph" (from Ralph Wiggum, persistent retry) + "-i" (inspection). Triggers: "ralphi", "사이클 점검", "loop budget", "cycle gate", "philosophy sweep", "between-cycle check"Votes: 0GitHub stars: 3
- PolishCompletion-claim gate. Before any "completed" / "done" / "passed" claim, output the self-check table. If ANY row is ✗, fix it first — do NOT report completion. Lean discipline-gate version (not the full quality-loop polish). Triggers: "polish", "완료 점검", "done check", "ready to ship", "PR ready", "완료 선언 전"Votes: 0GitHub stars: 3
- Refactoring DisciplineBehavior-preserving refactoring discipline. Activates when the task involves restructuring, extracting, or renaming existing code without adding new behavior.Votes: 0GitHub stars: 2
- InvestigateDeep investigation of bugs, performance issues, or unexpected behavior. TRIGGER THIS SKILL when users report bugs, errors, performance regressions, unexpected behavior, or ask to investigate, debug, diagnose, or find the root cause of any issue — whether they say "why is this broken", "investigate this bug", "find the root cause", "debug this issue", "this is failing and I don't know why", "performance degraded", or share error messages and stack traces. Do NOT trigger for feature requests, c...Votes: 0GitHub stars: 2
- DebugDebug and fix failing tests or errorsVotes: 0GitHub stars: 2