Review & quality
19,284 rows, by stacks then stars
1,766 of these skill files have been read, by 1,273 distinct authors, and what they tell an agent counted. What Review & quality authors agree on, and what they forbid
rjvim/ai-skills/skills/rjv-gated-build Skill
0★ repoUse automatically for financial/money, production-safety, security, destructive, high-blast-radius, zero-debt, multi-agent, or phase-reviewed builds where an INDEPENDENT reviewer must approve each slice—even when the user never says 'gated build'. Compose with `rjv-work-plan` and `rjv-spec-driven`: durable acceptance criteria live in `spec.md`; the plan holds cast, active criterion IDs, evidence, grill trail, and RESUME HERE. Enforce spec grill before code, ID-bound WORK/REVIEW, explicit APPROVED/REJECTED loops, crash/compaction durability, and model-economy routing. Triggers also include: gated build, grill each step, adversarial review, overnight run, high-stakes build, human QA gate, or repeated review rejections.
swaroopsm/skills/github-pr Skill
no license0★ repoOpen or update a GitHub pull request with an issue-number-prefixed title and a conventional-commits body, mirroring the git-commit style. Confirms before pushing an unpushed branch, defaults to a draft PR, and presents 3 title candidates for the user to pick from. When a PR already exists it refreshes only its managed section, preserving bot-generated and human-edited content. Use when the user says "pr", "open a pr", "create a pull request", "raise a pr", "update the pr", "update pr description", "refresh the pr", or asks to open, create, or update a PR for the current branch.
lankyjo/lankyjo-core-skills/skills/code/code-pipeline Skill
no license0★ repoMandatory post-write quality gate for every block of code written or changed. Chains code-review → simplify → verify → security-review so no code is called "done" until it is reviewed, simplified, run, and security-gated. Use whenever writing, editing, refactoring, or fixing code — automatically, not only when asked. (TDD is enforced separately, BEFORE you write code, by the /tdd pre-write gate — it is not a stage here.)
syniol/llm-software-project/.agent/skills/code-review Skill
0★ repoStep-by-step instructions for performing a structured code review, including the exact checklist, severity classification, and required output format.
jacobjustin8/skill-forge/examples/react-review Skill
0★ repoReview React + TypeScript code — components, hooks, a PR, or a pasted file — the way a senior frontend engineer would, prioritizing correctness, hooks bugs, performance, accessibility, and type safety over style nits. Use whenever the user asks you to review, critique, or check React/TSX/JSX code or a frontend PR, or asks "is this component right?" — even if they don't say "code review".
l3onlau/claude-life-admin/.claude/skills/visibility Skill
0★ repoSwitches the workspace between public mode (only framework files tracked — safe to publish) and private mode (life content versioned too — git history becomes memory). Use when the user wants to share or publish the repo, make it private, commit more of their content, or check whether the repo is safe to publish.
Enverge-Labs/skills/skills/worktree-new Skill
0★ repoGit worktree conventions for isolated parallel feature work. Use when starting a new coding task, feature, bugfix, or PR-review in a git repository — anytime work should get its own branch and worktree instead of touching the main checkout, or when the user asks to work in / create / list / remove a worktree. Applies to whatever repository Claude is currently in.
mlabarrere/dungeons-and-skills/skills/dnd-check Skill
no license0★ repoDungeons & Skills — a grounded D&D 2024 (5.5) character-builder skill pack: the AI reads a bundled rules catalogue + deterministic engine instead of hallucinating.
kalshamsi/fable-discipline-skills/evidence/cut-skills/reversibility-check Skill
0★ repoBefore executing any hard-to-reverse or destructive step, classify it as a one-way or two-way door; for one-way doors, substitute a reversible alternative or get explicit confirmation, and order work so mistakes can be undone. Use this whenever a plan or request involves deleting anything (files, records, branches, drafts), bulk or mass changes (renames, edits across many items, batch updates), sending to an audience (emails, announcements, messages to a group), publishing or deploying (posts, releases, reports going out), or changing shared systems others depend on. Trigger even when the destructive step is only one line of a larger task — "clean up and send" tasks are exactly where this matters most.
hsiangyilu/claude-design-skills/skills/screen-design-review Skill
0★ repo自建的畫面設計審查(design critique)技能。當使用者從 Claude Design、Figma 或其他原型工具拿到單一畫面、元件、原型,或一整條流程的所有畫面,需要嚴格、具體、可執行的視覺與可用性細節審查時使用。觸發情境:「幫我審這個畫面/這條流程的設計細節」、「critique 這個 mockup」、「這個原型的設計細節有什麼問題」、「跑一輪畫面設計審查」、「逐頁審每個畫面的細節」、「幫我挑視覺/對比/間距/字級的缺失」,或使用者貼上 Figma frame URL、原型 URL、設計截圖要求細節回饋時。即使使用者只說「看一下這個設計好不好」也應觸發。本技能聚焦畫面內的細節(type scale、spacing、對齊、色彩、對比、觸控目標、元件規範)+跨畫面的視覺一致性,強制具體性、拒絕空洞稱讚,目標是把設計品質推到「夠主觀、沒有設計細節缺失」為止。輸入是整條流程時,逐頁審細節並比對跨頁一致性。若需要審查的是跨畫面的流程邏輯本身(狀態覆蓋、返回路徑、分支對錯),改用 ux-flow-review。
MrToAster13/mrtoaster13-plugins/plugins/autonomous-workflows/skills/mine-provenance Skill
0★ repoAudit a skills/portfolio doc against ground-truth artifacts, then extend it. Verifies each claim with a git commit, file, or test result; sorts everything into VERIFIED, GAPS (claimed but unproven), and PROVENANCE-FLAG (evidence shows forked/derived). Transcripts are used only as discovery leads, never as proof. Never argues with the user's claims — it sorts them. Use when the user says "mine-provenance", "audit my portfolio", "verify my skills claims", "provenance check", or wants their portfolio backed by evidence.
NovaKepler513/claude-skills/obsidian-link-guard Skill
0★ repoObsidian 链接防断裂技能。用于文件重命名/迁移后,批量修复 wikilink、保留兼容入口、扫描全库断链并输出修复结果。 MANDATORY TRIGGERS: 改名后链接失效、红链、索引不到、空文件、404、断链、wikilink 修复、知识库链接体检、rename migration。 English triggers - "my wikilinks broke after renaming", "red links everywhere", "fix broken links in my vault", "zero-byte ghost files".
chsistrying/swift-ship-skills/skills/oss-readiness-audit Skill
0★ repoSystematic pre-publish audit for a project about to go public on GitHub (open source release or portfolio piece). Use when a repo is about to get its first commit/push, or when the user says things like "ready to publish", "open source my project", "push to GitHub first time", "portfolio repo review", "is this repo ready for GitHub", "can I make this public", or asks to check for secrets/leftover artifacts before sharing a repo. Walks a checklist covering git repo scope, .gitignore coverage, secrets/PII sweep, LICENSE, README completeness, misleading naming, internal process docs, doc consistency, and CI — gathering evidence with shell commands and producing a blocker/recommended/nice-to-have report. Language-agnostic across Swift, Node, Python, Rust, Go, etc. Trigger with "/oss-readiness-audit".
AmanKrSahu/zeroday-auditor/skills/code-quality-review Skill
0★ repoUse when the user wants a focused code-quality / maintainability review — 'review code quality', 'assess maintainability', 'find code smells', 'check SOLID/DRY adherence', 'is this codebase clean?', 'where's the dead code / duplication / bad error handling', 'refactoring opportunities', 'do we have enough tests?'. Produces a single evidence-backed report (code-quality-review.md) covering maintainability, consistency, smells, error handling, and test coverage. For a full multi-dimension audit use zeroday-auditor instead. Analysis only; never modifies source.
nubasu/Pseudo-Fable-Framework/.claude/skills/agent-framework-setup Skill
0★ repoGuided initial installation of the pseudo-fable framework family into a target project — inspect the target first, interview for a configuration, enforce the exclusivity rules, assemble CLAUDE.md / AGENTS.md / skills / hooks in the correct order with idempotent signature checks, then verify with agent-framework-doctor. Use when asked to install, set up, or add the agent framework (the pseudo-fable family, or any of its frameworks/modules) to a new or existing project.
nbsp1221/agent-skills/skills/commit Skill
no license0★ repoDetect the repo's commit convention (Conventional Commits, Gitmoji, or a custom template) and create commits. Use when asked to commit, write a commit message, stage and commit changes, or commit and push work.
gabriel-f-santos/coding-agents/skills/commit-message Skill
no license0★ repoGenerate a Conventional Commits message from staged changes. Use when the user asks to "write a commit message", "commit message", "gera a mensagem de commit", "commita isso", or right after staging with git add. Reads the staged diff, classifies it (feat/fix/docs/refactor/test/chore/...), and writes a <72-char imperative subject plus an optional body — it presents the message, it does not run the commit. Do not use for PR descriptions or changelogs — those are separate concerns.
The-Biomechanist/Windows-Toolchain-Forensics/mcp-main/skills/microsoft-code-reference Skill
0★ repoLook up Microsoft API references, find working code samples, and verify SDK code is correct. Use when working with Azure SDKs, .NET libraries, or Microsoft APIs—to find the right method, check parameters, get working examples, or troubleshoot errors. Catches hallucinated methods, wrong signatures, and deprecated patterns by querying official docs.
caoergou/erics-skills/skills/code-review Skill
0★ repoComprehensive code review combining Clean Code principles and senior engineer expertise. Reviews git changes for SOLID violations, security risks, clean code smells, and proposes actionable improvements.
Senior legal contract reviewer v1
evan66547/Contract-Reviewer-Agent-Eval/skills/senior-legal-contract-reviewer-v1 Skill
0★ repo高级法务合同审核智能体 v1.2,专注于商业合同风险管理,深度对齐中国《民法典》及相关司法解释,提供结构扫描、逻辑诊断、风险量化输出。
thedevappsecguy/sec-skills/skills/git-guard Skill
0★ repoSet up Claude Code hooks to block dangerous git commands before they execute, including `git push`, `git reset --hard`, `git clean`, `git branch -D`, and checkout or restore of the working tree. Use this skill when the user wants to prevent destructive git operations, protect a repo from accidental pushes or resets, add git safety hooks, or make git commands safer.
lu-zhengda/skill-port/test/fixtures/cursor-skill Skill
0★ repoReviews API changes for compatibility and naming consistency. Use when validating API contracts.
Verification before completion
winetree94/devtools/skills/verification-before-completion Skill
0★ repoVerify work before reporting completion. Use when finishing a task so the agent can identify validation steps, run relevant checks, confirm outcomes, and clearly report any remaining risk.
MasahiroSakoda/skills/skills/git-changelog Skill
0★ repoGenerate changelog from current branch changes. Use when the user asks to write a changelog entry, fill the changelog section of a PR, or determine if changes are user-facing.
mifunedev/skills/skills/ci-status Skill
0★ repoCheck the CI pipeline status for the current branch after pushing changes. Reports pass/fail with failure details. Use this after every push to confirm your changes are truly done — CI must be green. TRIGGER when: after git push, after committing changes, when asked to check CI, or when verifying that work is complete.
jiangyinzuo/agentic-software-engineering/commit-design-doc Skill
no license0★ repo基于当前 commit 或指定 commit range 生成代码设计文档,重点分析算法代码改动、代码重构改动,并输出结构化Markdown文档
sungkhum/agent-skills-pack/copy-review Skill
no license0★ repoClinical copy-editor that reviews text for communication issues. Use when the user asks to review prose, improve clarity, or polish written content without changing meaning.
garrettsiegel/lobstore/.github/skills/web-design-reviewer Skill
0★ repoThis skill enables visual inspection of websites running locally or remotely to identify and fix design issues. Triggers on requests like "review website design", "check the UI", "fix the layout", "find design problems". Detects issues with responsive design, accessibility, visual consistency, and layout breakage, then performs fixes at the source code level.
The same file is in 3 other repositories
yangshu2087/Codex/.agents/skills/architecture-decision-review Skill
no license0★ repoUse when a task involves architecture decisions, migrations, rollout planning, system boundaries, technical tradeoffs, or rollback risk.
twinfo-io/lifters-skills/skills/lf-commit Skill
no license0★ repoBuilds and creates a commit following conventional-commit style: type (feat/fix) resolved from the current branch prefix, ticket ID resolved from the branch name or from the project's issue tracker (Linear or GitHub Issues — detected from project docs or asked once per project), English subject and Portuguese body. Always presents the draft and only commits after explicit confirmation. Use when the user asks to commit, generate/review the commit message, or says something like 'commit this'/'cria o commit'. Suggests running /lf-push afterward.
synthetic-recon/synthetic-skills/product-design Skill
0★ repoDesigns and reviews production-quality user experiences for web and desktop applications with strong information hierarchy, interaction design, and state management. Use when building interactive apps, dashboards, admin panels, settings flows, multi-step workflows, or data-dense interfaces where the product needs to feel real and operational rather than decorative. Also use for UX reviews, critiques, and audits of existing product interfaces — "review my dashboard UX", "critique this flow", "audit this screen", or complaints about missing loading/empty/error states, modal overuse, or interfaces that feel like a demo. Other triggers include "make it feel like Linear", "production-quality UX", "real app feel", "good UX", or requests involving command palettes, keyboard shortcuts, inline editing, progressive disclosure, dense tables, or multi-screen flows. Pair with a visual design skill when the user also needs styling, theming, or brand expression.
taujiong/skills/skills/git-commit Skill
no license0★ repoGuide to automate the git commit workflow. Use when users want to commit changes, stage and commit, generate a commit message, or finalize an editing session with a commit. Also triggers when users mention commit my changes, help me commit, ok commit it, or looks good commit.
yaniv-golan/skill-packager-skill/skill-packager-workspace/iteration-1/eval-1-universal-repo/with_skill/outputs/proof-engine-skill/proof-engine/skills/proof-engine Skill
0★ repoCreate formal, verifiable proofs of claims with machine-checkable reasoning. Use when asked to prove, verify, fact-check, or rigorously establish whether a claim is true or false — mathematical, empirical, or mixed. Trigger phrases: "is it really true", "can you prove", "verify this", "fact-check this", "prove it", "show me the logic". Do NOT use for opinions, essays, or questions with no verifiable answer.
prof-ramos/skills/autoreview-agnostic/claude-code/.claude/skills/autoreview Skill
0★ repoCloseout structured code review. Use after non-trivial edits before commit/ship, or to review a local branch/PR/commit. The reviewer is read-only, returns a JSON findings bundle against a fixed schema, verifies every finding against real code, and rejects speculative/broad/unrealistic findings. Invoke via the /autoreview slash command or the autoreview subagent.
bensonmaxai/minis-coding-success-skills/skills/deployment-verification Skill
no license0★ repoUse when code works locally but you need to verify it will also work after deployment, packaging, or environment changes. Optimized for Minis on iOS: compare local vs target assumptions, check config/assets/startup paths, verify environment-sensitive behavior, and catch deployment-only failures before or after release.
HenryZhang-ZHY/agents/skills/conventional-git-commit Skill
0★ repoCreate git commits and draft, review, rewrite, or choose commit messages that follow Conventional Commits 1.0.0-beta.4. Use this skill whenever the user asks to commit changes, says "commit this", wants a commit message, asks whether a change is feat/fix/chore/refactor/etc., mentions breaking changes, changelogs, semantic versioning, release notes, squash commits, or wants staged changes summarized for git history. For exact specification wording and edge cases, consult references/conventional-commits-1.0.0-beta.4.md.
BenGardiner/ipynb-agent-skills/jupyter-notebook-linter Skill
0★ repoTool to format, lint, and validate Jupyter Notebook (.ipynb) files via the command line. Use when you need to resolve syntax errors, style violations, or reformat notebook code cells. Do not use for standard Python (.py) scripts or non-Python notebooks.
mralabs/claude-plugins/plugins/commit/skills/commit Skill
0★ repoCreate a conventional commit. Analyzes the current working tree diff and creates a commit with a conventional commit message. Supports --pr (push + open PR) and --merge/--m/--prm (PR + squash merge).
PeytonLi/skills/review-architecture Skill
0★ repoUse when you need a complete top-down architecture audit of any codebase, including full exploration of major paths, endpoint mapping, current-vs-target user journeys from documentation, and deep issue remediation plans with step-by-step fixes. Trigger phrases: architecture audit, full codebase walkthrough, map all endpoints, user journey analysis, current vs desired product flow, find and fix project issues, end-to-end architecture review.
Perdolique/workflow/skills/commit-creator Skill
0★ repoCreate English conventional commit messages and commits from the current changes. Use when the user wants to commit code, asks for a commit message, or needs monorepo scopes and dependency updates represented accurately.
bryanp/agent-skills/code/git/commit Skill
no license0★ repoCommit current work in git
Pratiyush/agent-catalog/skills/skill/data-validation Skill
0★ repoValidate CSV and JSON data files against schemas and quality rules. Use when the user asks to check data quality, validate a dataset, or verify file contents match an expected schema.
ThiagoPanini/aidriven/.claude/skills/python-project-setup Skill
0★ repoConfigure modern Python development tooling for any Python project — library, backend, CLI, or otherwise. Invoke this skill whenever a user wants to set up or improve Python tooling, configure pyproject.toml, add pre-commit hooks, set up linting or formatting, configure a type checker or test runner, or any request like "set up my Python project", "configure Python tooling", "add pre-commit to my project", "set up ruff and mypy", or "bootstrap Python dev tools". This skill works on both empty repositories and existing projects with partial configuration. It produces ONLY configuration files — no source code, no test logic.
rodrigoant/agent-skills/skills/commit Skill
0★ repoGeração de mensagens de commit padronizadas com emoji e tipo semântico
rawgroundbeef/skills/pr Skill
no license0★ repoWrite a human-readable PR title and description from the final diff (never from the conversation), open or update the pull request, and triage review feedback into address / defer / dismiss. Use when opening a PR, shipping a branch, tightening a PR title or description, or responding to review comments from humans or bots.
Access review coverage auditor
sisodiabhumca/agent-skills/skills/access-review-coverage-auditor Skill
0★ repoVendor-neutral skill to audit access review coverage and flag stale grants or missing attestations.
rynhardt-potgieter/sprint_workflow/plugins/sprint-workflow/skills/architecture-drift-check Skill
0★ repoDetect architectural drift and erosion by comparing implemented or planned work against the project's Linear "Architecture & Roadmap" document. Use this skill in /sprint-plan and /sprint-enrich (planning-time drift), in /sprint-start Phase 3 QA and code review (implementation-time drift and erosion), and in /sprint-retro (sprint-level drift summary). Defines drift vs erosion, the reflexion-modeling comparison method, severity rules, the standard report format, and the conditions under which the check is skipped gracefully.
eunai/skills/.agents/skills/commit-standards Skill
0★ repoAuthor and validate git commit messages against Conventional Commits, layering any repo-local rules on top. Use when about to commit, when writing or fixing a commit message, or when the user invokes /commit-standards or asks about commit message conventions. Not for changelog entries (use the changelog skill) or for browsing git history.
qiaoruntao/skills/commit Skill
no license0★ repoCommits code changes to a git repository using a Haiku subagent. Use this skill whenever the user wants to commit — phrases like "let's commit", "commit this", "commit the auth changes", "go ahead and commit", "time to commit", "push a commit", "save my changes to git". Also triggers when the user finishes describing a change and says something like "ok ship it" or "done, commit it". The skill handles staging the right files based on intent, updating .gitignore for files that shouldn't be tracked, and writing a commit message — the user doesn't need to think about any of that.
AnotherDaniel/tsfagent/.agents/skills/quality-reviewer Skill
0★ repoReview overall project quality: TSF graph coverage and content, OFT tracing gaps, evidence relevance, code best practices, and CI structure. Use when: auditing quality artifacts, reviewing traceability completeness, checking TSF graph health, assessing evidence quality, verifying project structure and CI setup. Keywords: review, audit, quality, coverage, gap, TSF graph, OFT trace, evidence, best practices, CI, lint.
mtzanidakis/skills/skills/github/github-merge-prs Skill
0★ repoUse when the user wants to bulk-merge open GitHub pull requests on the current repo. Triggers include "merge open PRs", "merge the dependabot PRs", "clean up open PRs", "merge all PRs", "merge what's mergeable". Handles Dependabot conflicts by commenting `@dependabot rebase` (never closing the PR), polls for the rebased commit, and re-attempts the merge. Surfaces (does not bypass) failing checks or required reviews.
frankenmason/amm-skills/amm-save-branch Skill
0★ repoPersist a mid-session branch point (topic shift) to long-term memory, marking where the agent pivoted from one scope to another so the prior scope can be resumed later. Trigger when the user steers to a new task while prior work is incomplete, or when the agent self-detects a scope change worth bookmarking.
MAHDTech/agent-skills/skills/authoring/skill-audit Skill
0★ repoPeriodically audit the whole skill collection for health — validate each skill's frontmatter, clarity, and category, verify every cross-reference resolves, and surface duplicates, conflicts, retirement candidates, and missing-skill gaps. Use when you want to audit or health-check the skills, spring-clean the collection, confirm the router and cross-references are accurate, or find skills to merge, split, retire, or create. Hands findings to /skill-creator to fix or create and /deprecate-skill to retire.
Agents-On-Rails/aor-skills/.claude/skills/aor-review-adhoc Skill
0★ repoSpawn an ad-hoc SME on demand for a one-off review perspective. Optionally persist the SME for reuse.
Montesplenic117/CC/src/skills/bundled/verify Skill
no license0★ repoRebuild Claude Code CLI source from source maps into runnable TypeScript for study, debugging, and local development
gerardbalaoro/skills/skills/engineering/compose-commits Skill
no license0★ repoCompose uncommitted changes, a linear commit range, branch, or PR into logical commits. Use when the user wants to split, clean up, organize, or rewrite commits. Always inspect first, propose a breakdown, get one confirmation, create a backup branch, then rewrite.
fbsmna-coder/karpathy-pro-max/skills/hostile-self-review Skill
0★ repoReview your own diff as a skeptic before declaring done. Use after implementing any non-trivial change to catch bugs, scope creep, hidden assumptions, and overclaims that you would otherwise ship.
eooo-io/ai-native-engineering-doctrine/04-agent-skills/code-review Skill
0★ repoReview code for correctness, maintainability, security, performance, and test quality.
Grill with docs pre prune 3728a775
stevennitesh/programming-agent-skills/docs/validation/evals/grill-with-docs-pre-prune-3728a775 Skill
0★ repoCompose one explicitly invoked repo-backed decision through $grilling and $domain-modeling; return intact component results without downstream execution.
hansonkim/agent-skills/skills/pr-understanding-question Skill
0★ repoGenerates context-aware understanding questions from a PR, design notes, and code changes to verify whether the reviewer truly understands the why, trade-offs, risks, and behavior of the changes before approving. Runs as an interactive conversation: presents questions, waits for the reviewer's answers, then evaluates each answer against the PR facts. Use this skill whenever: - A reviewer wants to verify their own understanding of a PR before approving it - A user shares a PR (link, diff, description, or code changes) and wants understanding questions generated - A user says "check my understanding of this PR", "generate questions for this PR", "quiz me on this change" - A user wants to evaluate whether their answers about a PR are accurate and complete - A code review process needs structured questions to probe design decisions, risks, or trade-offs Do NOT trigger for: general code review feedback, summarizing a PR, writing a PR description.