Review & quality
19,016 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
Appstore push testflight build
jimmynotjames/simple-recurring-budgets/.claude/skills/appstore-push-testflight-build Skill
no license0★ repoBuild and ship a TestFlight build end-to-end — creates a release branch, runs `fastlane ios beta` (auto-bumps build number from ASC, archives, uploads, auto-commits the pbxproj bump), then pushes, opens a chore PR, waits for CI, and squash-merges. Use when cutting a new TestFlight build. Invoked via /appstore:push-testflight-build.
jimmynotjames/simple-recurring-budgets/.claude/skills/create-pr Skill
no license0★ repoTurn already-written code changes into a reviewed, gated, open PR. Spawns a fresh-context code-reviewer subagent for a correctness-first look at the diff, audits the AGENTS.md cross-cutting concerns checklist (accessibility, localization, translations, Mixpanel, UI test screen objects), fills in any gaps autonomously, runs `make gate`, commits residual fixes, pushes the branch, and opens the PR. Use when the solution is mostly done and you want to review, gate, and ship it to a PR — without involving a GitHub issue. Does NOT merge; pair with /merge-pr after review. Invoked via /create-pr.
jimmynotjames/simple-recurring-budgets/.claude/skills/openspec-verify-change Skill
no license0★ repoVerify implementation matches change artifacts. Use when the user wants to validate that implementation is complete, correct, and coherent before archiving.
Translation accessibility size check
jimmynotjames/simple-recurring-budgets/.claude/skills/translation-accessibility-size-check Skill
no license0★ repoAd-hoc visual check for localized truncation / overflow / RTL-mirroring at large Dynamic Type. Renders the app in 7 locales at forced .xxxLarge, captures screenshots of the key screens, and has Claude eyeball them. Run on demand (e.g., before an App Store submission) — NOT part of make test / CI / pre-commit. Use when the user asks to check localized layouts, translation truncation, RTL, or large-text rendering before release.
jimmynotjames/simple-recurring-budgets/.cursor/skills/app-store-review Skill
no license0★ repoPrepare for App Store review and prevent rejections. Covers App Store review guidelines, app rejection reasons, PrivacyInfo.xcprivacy privacy manifest requirements, required API reason codes, in-app purchase IAP and StoreKit rules, App Store Guidelines compliance, ATT App Tracking Transparency, EU DMA Digital Markets Act, HIG compliance checklist, app submission preparation, review preparation, metadata requirements, entitlements, widgets, and Live Activities review rules. Use when preparing for App Store submission, fixing rejection reasons, auditing privacy manifests, implementing ATT consent flow, configuring StoreKit IAP, or checking HIG compliance.
jimmynotjames/simple-recurring-budgets/.cursor/skills/appstore-push-release-build Skill
no license0★ repoBuild and upload the App Store release-candidate build — confirms the release checklist has been worked through, prompts for the new App Store version number, bumps MARKETING_VERSION on a release branch, then runs `fastlane ios beta` so the candidate lands in TestFlight (auto-bumps build number from ASC, archives, uploads, auto-commits the pbxproj bump), then pushes, opens a chore PR, waits for CI, and squash-merges. The same binary is then manually tested via TestFlight and attached to the App Store version in ASC — one build serves both. Does NOT submit for review — that stays a manual App Store Connect step. Use when the release checklist reaches the release-candidate build step, or when asked to push a release build. Invoked via /appstore:push-release-build.
jimmynotjames/simple-recurring-budgets/.cursor/skills/create-pr Skill
no license0★ repoTurn already-written code changes into a reviewed, gated, open PR. Spawns a fresh-context code-reviewer subagent for a correctness-first look at the diff, audits the AGENTS.md cross-cutting concerns checklist (accessibility, localization, translations, Mixpanel, UI test screen objects), fills in any gaps autonomously, runs `make gate`, commits residual fixes, pushes the branch, and opens the PR. Use when the solution is mostly done and you want to review, gate, and ship it to a PR — without involving a GitHub issue. Does NOT merge; pair with /merge-pr after review. Invoked via /create-pr.
jimmynotjames/simple-recurring-budgets/.cursor/skills/openspec-verify-change Skill
no license0★ repoVerify implementation matches change artifacts. Use when the user wants to validate that implementation is complete, correct, and coherent before archiving.
Translation accessibility size check
jimmynotjames/simple-recurring-budgets/.cursor/skills/translation-accessibility-size-check Skill
no license0★ repoAd-hoc visual check for localized truncation / overflow / RTL-mirroring at large Dynamic Type. Renders the app in 7 locales at forced .xxxLarge, captures screenshots of the key screens, and has Claude eyeball them. Run on demand (e.g., before an App Store submission) — NOT part of make test / CI / pre-commit. Use when the user asks to check localized layouts, translation truncation, RTL, or large-text rendering before release.
hsnuhow/how-skills/plugins/ponytail/skills/ponytail-review Skill
no license0★ repoCode review focused exclusively on over-engineering. Finds what to delete: reinvented standard library, unneeded dependencies, speculative abstractions, dead flexibility. One line per finding: location, what to cut, what replaces it. Use when the user says "review for over-engineering", "what can we delete", "is this over-engineered", "simplify review", or invokes /ponytail-review. Complements correctness-focused review, this one only hunts complexity.
The same file is in 3 other repositories
wjb127/claude-skills/verify-pixels/skills/verify-pixels Skill
0★ repoVerify that ad and analytics pixels actually fire — GA4, Meta, GTM, TikTok, Naver, Kakao — by driving a real browser, then produce a screenshot + network evidence package. Includes an auto-fix loop for the common install errors. 광고 픽셀·GA4 설치 검증 + 클라이언트 제출용 증거 패키지 생성.
Studio-Moser/skills-n-stuff/plugins/pm/skills/codex-computer-use Skill
0★ repoAsk the Codex CLI (running whatever model it's configured with) to run local app verification that needs computer use, browser automation, simulators, screenshots, app launching, or independent runtime inspection. This is how the Codex model is invoked for computer-use work. Use when the user asks Claude to test a flow, verify UI behavior, inspect a running app, capture screenshots, or report confirmation and feedback about implemented behavior that benefits from computer use functionality.
Studio-Moser/skills-n-stuff/plugins/pm/skills/codex-review Skill
0★ repoAsk the Codex CLI (running whatever model it's configured with) for an independent code review of uncommitted changes, a branch diff, a commit, or a specific implementation. This is how the Codex model is invoked for review work. Use when the user asks Claude to have Codex review work, when the model-selection rubric calls for a cross-vendor review perspective, or when Codex should audit a diff, find bugs or regressions, or compare Claude's implementation against requirements. For a review by Claude itself, use the normal review process instead.
Studio-Moser/skills-n-stuff/plugins/pm/skills/house-rules Skill
0★ repoUse when making any code change and you need Studio Moser conventions for branch naming, file/documentation naming, commit messages, PR format, testing discipline, or pre-commit security checks, or when you just need a quick convention lookup. Loaded by pm:dev-task and pm:sprint-dev.
Evan-Daruwalla/claude-skill-suite/code-check Skill
0★ repoPost-write verification pass on code you just wrote or edited: re-read the real diff, make it actually run, confirm non-trivial logic has one runnable check behind it, and confirm every changed line traces to the request. Fires from the PostToolUse hook after Edit/Write on a code file, and on "check this", "verify that code", "did that actually run", "code-check". Not a full audit (/audit) and not a bug hunt (/code-review) — this is the did-what-I-just-wrote-actually-work pass.
Evan-Daruwalla/claude-skill-suite/commit-gate Skill
0★ repoBlocks commits that stage a secret, and helps produce clean atomic commits. The gate is TWO deterministic hooks over the shared pm-secretscan scanner: a native git pre-commit hook (covers shell commits) and a PreToolUse hook (covers commits the model makes via Bash) — because a PreToolUse hook alone misses shell commits. Use when the user says "commit-gate", "guard my commits", "block secret commits", "set up the pre-commit hook", or when splitting a large change into atomic commits / writing commit messages. Reads .claude/secrets-inventory.md.
Evan-Daruwalla/claude-skill-suite/determinism-guard Skill
0★ repoEPHEMERAL invariance checker — run a command repeatedly in one shot and prove it produces the SAME output every time. Three checks over one engine: default runs `--cmd` N times and compares stdout + exit code byte-exact across runs (first-divergence line diff on failure); `--files "a,b,c"` ALSO sha256s each listed file after every run to catch non-reproducible rebuilds; `--shuffle-stdin <file>` runs the command twice — input as-is vs. its lines shuffled by a fixed-seed PRNG — to catch order-dependence. No baselines are stored (freeze an output across time with golden-lock instead). Use when the user says "is this deterministic", "does it give the same output every run", "check for flakiness", "reproducible build check", "is my output order-dependent", "determinism-guard", or wants a command proven stable before it is frozen or scheduled. Deterministic, zero dependencies, no model calls.
Evan-Daruwalla/claude-skill-suite/etl-validate Skill
0★ repoRead-only source-vs-target assertion after a data transform or copy — prove every row actually moved. Compares a CSV file or a SQLite table against another (--src / --dst as csv:<path> | sqlite:<db>:<table>) on two axes: row counts equal, and a content checksum equal (each row's selected columns tab-joined, sha256'd, then XOR-combined so the result is order-independent by construction). Optional --key <col> names the src keys missing from dst (first 10), turning a bare mismatch into named rows. Use when the user says "validate the copy", "did every row move", "verify the ETL", "source vs target check", "reconcile the table after a rebuild", or after a cache/table rebuild. SQLite opens mode=ro. Deterministic, Python stdlib only, no model calls.
Evan-Daruwalla/claude-skill-suite/experiment-log Skill
0★ repoReproducibility provenance for a single run — runs a command and appends ONE JSON line capturing exactly what produced the result: ISO timestamp, cwd, cmd, exit code, duration, git {commit, dirty}, tool versions {node, python}, and the sha256 of every declared input (hashed before the run) and output (after), plus a note. Append-only JSONL; never rewrites the log. Use when the user says "log this run", "record provenance", "experiment log", "make this run reproducible", "what produced this output", "capture inputs/versions for this backtest", or wants a run pinned so it can be reproduced. Machine provenance only — it never touches HANDOFF.md or the project record (project-memory owns the narrative). Deterministic, zero dependencies, no model calls.
Evan-Daruwalla/claude-skill-suite/github-repo-polish Skill
0★ repoMake an EXISTING GitHub repo professional: its PRESENTATION (name, one-line description, topics, README structure, homepage, semver tags/releases) AND a clean git BRANCH WORKFLOW (when to branch, GitHub Flow: feature branch → PR → merge → delete) — using grounded `gh`/`git` commands, propose-then-confirm on every public change. Use when explicitly asked to set up/professionalize a repo's presentation (README/description/topics/releases) OR for branching / pull-request / GitHub-Flow guidance. Do NOT use for writing code, routine "commit this"/"push this" of already-settled work, creating a new repo/ scaffolding, or CI/Actions/Projects/Wikis.
Evan-Daruwalla/claude-skill-suite/golden-lock Skill
0★ repoFreeze ANY output as a byte-exact golden baseline and diff on change — one engine over three inputs: a command's stdout, a fixture file, or a prompt/text asset. "freeze" records the baseline (output.txt + meta.json under .golden/); "check" re-produces it and fails with a line-numbered diff on any drift, comparing bytes AND exit code. Generalizes frozen-regression-test discipline (an output pinned byte-exact) to arbitrary outputs. Use when the user says "freeze this output", "golden test", "lock the baseline", "snapshot freeze", "did the output change", "prompt regression", "guard against output drift", or wants a fixture/prompt/command pinned so any change is a failing diff. Deterministic, zero dependencies, no model calls.
Evan-Daruwalla/claude-skill-suite/history-leak-scan Skill
0★ repoDeterministic secret scanner for git repositories — no dependencies, no API key. Scans full git history (every version ever committed) or the staged diff for leaked credentials: per-provider keys (AWS, GitHub incl. fine-grained PATs, Slack, Google, Stripe, Alpaca, Anthropic, OpenAI, npm, HuggingFace, SendGrid, Twilio, Resend), private-key blocks, JWTs, sensitive filenames (.env, .pem, id_rsa, *_keys.env), high-entropy assignments, and weak/short passwords. Use when the user says "scan for secrets", "leak scan", "check for leaked keys/credentials", "did I commit a secret", "history-leak-scan", or after any repo goes public / any suspected exposure. The same scanner backs commit-gate (staged mode). Reads .claude/secrets-inventory.md for what credentials exist.
Evan-Daruwalla/claude-skill-suite/opus-workers Skill
0★ repoWhen an expensive model is orchestrating (e.g. Claude Fable 5 / Opus at high effort) and you're about to spin up subagents or a workflow for non-trivial work, run the WORKER agents on a cheaper tier — one tier down (e.g. Opus) at medium/high effort for substantive generation, two tiers down (e.g. Sonnet) at low/medium for mechanical bulk — after pre-registering a review rubric, then have the orchestrator review each output against that rubric: accept + present, or send back with specific pointers. Cheaper workers do the bulk generation; the orchestrator stays a thin reviewer. Skip for trivial one-shot tasks (inline), for GATED actions (commit/push/publishing stay with the orchestrator), and when no cheaper tier fits the task.
Evan-Daruwalla/claude-skill-suite/seed-control Skill
0★ repoRead-only static scan for UNSEEDED randomness — the reproducibility guard. Flags Python random.<fn>() in any file with no random.seed() anywhere in it, np.random.<fn>() with no np.random.seed()/default_rng() in the file, and every JS/TS Math.random() (no seed API exists). Reports file:line:snippet, exit 1 on findings. Suppress a known-fine line with a `# seed-ok` / `// seed-ok` comment. Use when the user says "scan for unseeded randomness", "reproducibility check", "did I forget to seed", "seed-control", or before pinning a frozen/regression run whose output must be byte-stable. Deterministic, zero dependencies, no model calls, writes nothing.
Evan-Daruwalla/claude-skill-suite/shell-portability Skill
0★ repoRead-only SYNTAX/semantics scanner for cross-shell traps in shell scripts — the proactive counterpart to your own reactive troubleshooting runbook. Flags, with file:line + why it breaks + the PowerShell-5.1-safe fix: in .ps1/.psm1, the && / || pipeline chain (PS 5.1 parser error), ?: ternary, ?. null-conditional and ?? / ??= null-coalescing (PS7-only), Read-Host / pause / Out-GridView (block a non-interactive or scheduled run), Set-Content / Add-Content / Out-File written WITHOUT -Encoding (ANSI/UTF-16 default corrupts UTF-8), and bash-style NAME=value assignments; in .sh/.bash, PowerShell-isms ($env:NAME, Verb-Noun cmdlet calls). Use when the user says "shell-portability", "scan for PS5.1 traps", "will this script run on PowerShell 5.1", "check my .ps1 for portability", "lint my shell scripts", or before shipping a script to the scheduled-task / CI path. Encoding and filename quirks belong to path-quirk-audit; this is syntax only. Deterministic, zero dependencies, no model calls.
Evan-Daruwalla/claude-skill-suite/trusted-advisor Skill
0★ repoCandid advisor, two layers. BASELINE, all interactions: never a yes-man; verdict first; no hollow affirmations; honest calibration. FULL-CRITIQUE MODE when the user wants critique, devil's advocate, or unfiltered feedback — "be honest", "don't sugarcoat", "poke holes", "tell me if I'm wrong", "challenge me", "what am I missing", "critique this", or similar — or proactively, but ONLY when the flaw is significant (wasted money/time, security, data loss, hard to reverse) AND staying silent would make the outcome worse; minor issues get one sentence and the task proceeds, never an unsolicited review. Yields to project/task instructions on format and tone; never on honesty.
Evan-Daruwalla/claude-skill-suite/venue-fit Skill
0★ repoScore a draft against ONE venue's real published reviewer criteria before submitting — per-criterion verdict with evidence quoted from the draft, plus the concrete gap for anything short. Refuses to emit an acceptance probability. Learns only from REAL post-submission reviewer feedback, stored as data. Use when the user says "venue-fit", "would this get into <venue>", "review this against <venue/workshop>", "am I ready to submit", or names a submission target. Must be invoked — never fires on its own.
dkmqflx/claude-tools/.claude/skills/aws-cloudformation Skill
no license0★ repoAuthor, validate, and troubleshoot AWS CloudFormation templates. Covers template authoring with secure defaults, pre-deployment validation (cfn-lint, cfn-guard, change sets), and root-cause diagnosis of failed stacks using CloudFormation events and CloudTrail correlation.
dkmqflx/claude-tools/.claude/skills/branch-done Skill
no license0★ repoDelete the current feature branch and switch back to the base branch after opening a PR. Use when the user has just opened a PR and wants to clean up — phrases like "PR 올렸으니 정리해줘", "브랜치 정리해줘", "정리하고 돌아가줘", "clean up this branch", "delete feature branch", "/branch-done". Do NOT use inside a git worktree — that is the worktree-done skill.
The same file is in 1 other repository
dkmqflx/claude-tools/.claude/skills/commit-each Skill
no license0★ repoSplit the current git changes (working tree + staged) into one commit per logical unit of work. Use when the user asks to commit changes separately by task — phrases like "각각 커밋해줘", "작업별로 커밋", "변경사항 분리해서 커밋", "split commits", "commit by task", "/commit-each".
dkmqflx/claude-tools/.claude/skills/grill-with-docs Skill
no license0★ repoGrilling session that challenges your plan against the existing domain model, sharpens terminology, and updates documentation (CONTEXT.md, ADRs) inline as decisions crystallise. Use when user wants to stress-test a plan against their project's language and documented decisions.
The same file is in 8 other repositories
dkmqflx/claude-tools/.claude/skills/nestjs-validation-pipes Skill
no license0★ repoNestJS validation and pipes best practices. Use when validating or transforming request input in NestJS — DTO validation, ValidationPipe, class-validator/class-transformer, or custom pipes. Triggers on ValidationPipe, @IsString/@IsEmail, ParseIntPipe, whitelist, or transform.
The same file is in 1 other repository
dkmqflx/claude-tools/.claude/skills/open-pr Skill
no license0★ repoPush the current branch and open a pull request against the detected base branch, with an AI-drafted title and body confirmed by the user before creating. Use when the user wants to open, create, or submit a PR — phrases like "PR 올려줘", "PR 열어줘", "PR 만들어줘", "open a PR", "create a pull request", "submit for review", "/open-pr". Do NOT use to review, comment on, or update an existing PR.
The same file is in 1 other repository
dkmqflx/claude-tools/.claude/skills/start-branch-from-dev Skill
no license0★ repoSync the local dev branch to the latest remote, then create a new well-named branch from it and switch to it. Use when the user wants to start fresh work based on dev — phrases like "dev에서 브랜치 만들어줘", "dev 기준으로 작업 시작", "dev 기반 브랜치", "start from dev", "branch off dev", "/start-branch-from-dev". Do NOT use when branching from the current position is fine — use /start-branch for that.
dkmqflx/claude-tools/.claude/skills/start-branch Skill
no license0★ repoCreate a new git branch with an appropriate name for the current task and switch to it immediately — no new worktree directory, just a branch. Use when the user wants to start work on a fresh branch from the current position — phrases like "브랜치 만들어줘", "새 브랜치에서 작업 시작", "start a new branch", "create a branch for this", "작업 브랜치 파줘", "/start-branch". Do NOT use when the user explicitly wants a worktree (separate directory).
dkmqflx/claude-tools/.claude/skills/worktree-done Skill
no license0★ repoRemove the current git worktree, switch the main checkout to the base branch, and delete the feature branch. Use when the user has finished work started with the worktree skill (or `/worktree`) and wants to clean up — phrases like "워크트리 정리해줘", "워크트리 삭제해줘", "clean up this worktree", "remove worktree", "/worktree-done". Do NOT use from the main checkout (not a worktree) — that is the branch-done skill.
The same file is in 1 other repository
dkmqflx/claude-tools/.claude/skills/worktree Skill
no license0★ repoCreate a new git worktree on a fresh branch and switch into it before starting work. Use when the user asks to start a task in a new worktree, work in isolation from the main checkout, or invokes "/worktree" — phrases like "워크트리 만들어줘", "새 브랜치 파서 작업해줘", "worktree에서 작업 시작". Do NOT use for removing or cleaning up an existing worktree.
The same file is in 1 other repository
cloverink/shipwright/skills/merge-pr Skill
0★ repoMerge current PR after CI passes. Waits for all checks, squash merges, deletes branch, switches to main.
cloverink/shipwright/skills/review-code-fix Skill
0★ repoAutomated code review → fix → re-review loop. Runs until score >9 or max 3 rounds. Fixes Critical/Major/Warnings only — Suggestions are deferred.
cloverink/shipwright/skills/review-docs Skill
0★ repoVerify Phase D doc updates are accurate. Binary pass/fail gate — checks stat accuracy, file path validity, code-doc sync, completeness. Runs after /update-docs.
cloverink/shipwright/skills/review-full Skill
0★ repoSmart review dispatcher. Auto-detects scope from diff (frontend/backend/docs/both), runs the right fix loops until all gates pass. Used as Phase R entry point in /ship.
cloverink/shipwright/skills/review-ux-fix Skill
0★ repoAutomated UX review → fix → re-review loop. Target is perfection (10/10). Fixes ALL severities including Suggestions — because UX polish compounds.
prvthmpcypher/skills-design/skills/design-critique Skill
0★ repoYou are a senior UI/UX designer providing structured, actionable design critiques. When given a design description, evaluate it across multiple dimensions and provide specific, constructive feedback. ## Process 1. Analyze the design across key UX dimensions 2. Evaluate visual hierarchy and layout 3. Assess typography, color, and spacing 4. Check accessibility compliance 5. Provide specific, actionable recommendations ## Output Format ## Design Critique: \[Design Name\] ### Overall Score: \[X\]/10 ### Visual Hierarchy Strengths: What works well Issues: What needs improvement Recommendation: Specific fix ### Typography - Font pairing assessment - Readability score - Scale consistency ### Color & Contrast - Palette harmony - WCAG compliance - Emotional resonance ### Layout & Spacing - Grid alignment - White space usage - Component balance ### Accessibility - Color contrast ratios -...
agony1997/TouchFish-Skills/plugins/git-nanny/skills/git-nanny Skill
0★ repoGit 全方位專家:涵蓋 Commit 訊息生成、Pull Request 建立與審查、分支策略管理、版本發布與 Changelog。專精 Conventional Commits、Code Review、Git Flow / Trunk-Based / GitHub Flow、Semantic Versioning、Release Notes 撰寫。關鍵字: commit, pull request, PR, branch, release, changelog, git flow, trunk-based, github flow, semantic versioning, conventional commits, code review, git tag, 提交, 分支, 版本發布, 合併請求, 程式碼審查, 更新日誌, 版本號, semver, hotfix, 緊急修復, merge, 合併, 合併策略, rebase, squash, tag, 標籤
The same file is in 2 other repositories
agony1997/TouchFish-Skills/plugins/reviewer/skills/reviewer Skill
0★ repo專案規範審查與萃取:讀取專案內的規範文件執行合規審查,或從現有程式碼反向萃取隱含慣例產出 .standards/ 草稿。 規範文件由使用者維護在專案中(如 .standards/ 目錄),skill 提供審查與萃取兩種工作流。 使用時機:實作完成後要求審查、CI 前檢查、程式碼合規確認、規範檢查、從程式碼萃取慣例。 關鍵字:review, 審查, 規範, standards, compliance, 合規, 檢查, CI, pre-commit, code review, 程式碼審查, 規格檢查, lint, linting, coding standards, 編碼規範, 合規審查, 規範審查, quality, 品質, 程式碼品質, extract, 萃取, generate standards, 產出規範, 慣例分析。
The same file is in 2 other repositories
ArthurZakirov/SystemSmith/skills/self-review Skill
0★ repoAnalyze the current conversation or mine historical chat sessions for correction patterns, wrong conclusions, or unnecessary user intervention. Use when the user wants to improve prompts, skills, workflows, or guardrails based on repeated mistakes.
AI-Driven-School/aiki/.claude/skills/review Skill
0★ repoRun code review against design documents, checking acceptance criteria, security, performance, and test coverage. Use when reviewing implementation quality or running /review.
jgsystemsconsulting/jgs-goal-spec/skills/spec-review Skill
0★ repoAutonomously harden an engineering specification through a roundtable of independent engineering-discipline reviewers (architecture, security, test/QA, operations, product). Each reviewer is a real subagent thinking independently; the orchestrator collects findings, revises the spec itself, and loops until convergence (no CRITICAL or MAJOR findings) with NO user checkpoints. Use when you say "review this spec", "harden this spec", "spec review", "run the roundtable on this spec", or want a spec battle-tested before implementation. Self-contained, with no dependency on any other skill.
Homepage copy review checklist
ArthurZakirov/ProofStack/skills/homepage-copy-review-checklist Skill
no license0★ repoReview homepage copy for clarity, authority, reader value, and layout usefulness.
ArthurZakirov/ProofStack/skills/linkedin_content_validation Skill
no license0★ repoUse when drafting, saving, formatting, validating, or repairing LinkedIn profile and content assets in this repository, including headlines, About sections, Experience descriptions, project copy, posts, articles, and LinkedIn image specs.
latuconsinafr/claude-code-config/skills/review Skill
no license0★ repoUse before committing or opening a pull request — reviews staged changes as a principal engineer for correctness, edge cases, and logic flaws. Always invoke this before creating a PR.
latuconsinafr/claude-code-config/skills/test-cases Skill
no license0★ repoUse when you need to document E2E test cases for a feature or bug fix that can't be covered by unit or integration tests alone. Auto-detects ticket from branch name. Reuses codebase context from /plan if already run; falls back to git diff + explorer when invoked standalone. Outputs to a markdown file by default.
Jewel591/claude-with-codex/skills/using-codex-collaboration Skill
0★ repoUse proactively whenever the work involves substantial code changes (multi-file, core logic, architecture), a freshly written execution plan, a sticky bug that resists single-perspective diagnosis, or a non-obvious design decision that needs adversarial challenge — establishes when and how to bring Codex in as a reviewer (not as an executor). The ONLY channel is the persistent Codex pane reached via tmux — this holds even when Claude's own session is NOT running inside tmux (drive the tmux server externally). Direct `codex` CLI invocation (codex exec etc.) is FORBIDDEN unless the user explicitly requests it. Covers the bundled codex-pane.sh helper (find-or-launch, send, wait, capture, cleanup), the predefined collaboration scenarios (change review, plan review, adversarial challenge, bug rescue), and the review⇄fix consensus loop that runs until a clean verdict.
az9713/youtube-transcript-to-skills/.claude/skills/setup-validation Skill
no license0★ repoSet up a build/test/lint validation loop for the current project. Detects the tech stack and configures CLAUDE.md with verification commands.
Reporting artefact detection before claiming anomaly
Ed3Design/ed3design-skill-bundles/async-forensik/skills/reporting-artefact-detection-before-claiming-anomaly Skill
0★ repoUse when observing a "surprising anomaly" in backtest output, SQL query result, multi-run-eval-report, ML-model-performance-table, or any reporting-layer artefact suggesting a system bug, market outlier, methodological problem, or drift. Before treating the anomaly as real and dispatching forensic resources, run a 3-step triage: (1) NULL-handling check — are NULLs in numerator/denominator mishandled, was IS NOT NULL omitted? (2) Cross-window / unique-ID check — do overlapping sub-windows / joins count rows multiple times? (3) Methodology-consistency check vs previous report — was the same SQL replicated exactly? Trigger phrases like "why is WR suddenly so low", "I don't understand this anomaly", "24h drift", "baseline has changed", "backtest gives different numbers today than yesterday". Do NOT load for confirmed-real anomalies, first-time-setup-debugging without baseline, code-bugs in the algorithm itself, or UI rendering bugs.
Aggregate code review after tdd tasks
Ed3Design/ed3design-skill-bundles/code-quality/skills/aggregate-code-review-after-tdd-tasks Skill
0★ repoUse when implementing a multi-task TDD plan via `superpowers:subagent-driven-development` AND all individual tasks have already passed their per-task two-stage review. Per-task reviews check one commit against one task — they cannot catch cross-commit drift that only appears when reading the full feature end-to-end. Triggers a final aggregate review looking for: state-machine/algorithm-order drift between spec and code, naming-inconsistency across commits, end-to-end data-flow integrity, commit-message honesty, pre-commit-hook coverage gaps (lazy imports), test-suite cohesion. Trigger phrases like "Tasks 1-N all done", "aggregate review before live-run", "cross-commit drift check", "verify end-to-end behavior against spec". Do NOT load for single-commit reviews, per-task reviews within SDD, branches with <3 atomic feature-commits, or hotfixes without spec-text target.
Code review backlog cost warning
Ed3Design/ed3design-skill-bundles/code-quality/skills/code-review-backlog-cost-warning Skill
0★ repoUse when the user/agent is about to push code, merge a feature branch, or claim work as "done" AND no code-review has happened recently. Specifically trigger when ANY hold: >7 days since last `requesting-code-review` invocation, >30 atomic commits accumulated since last review, >5000 LoC changed, pre-push-hook code-review-warning bypassed ≥2 times in succession, or the user says "haven't done a review in weeks", "I'll just push this", "the hook is only a warning, ignore it", "it's just a small fix on top". STOP and surface a cost-estimate-warning BEFORE the push — quantify expected backlog cost (wallclock + tokens + re-review cycles) and offer chunked-parallel-review-dispatch as alternative. Do NOT load for fresh branches (≤3 commits, ≤500 LoC, ≤2 days), explicitly-marked WIP pushes, inside an active review session, or one-line typo/doc commits.
Ed3Design/ed3design-skill-bundles/code-quality/skills/code-review-chunk-dispatch Skill
0★ repoUse when the code-review-backlog is large (>30 commits OR >5,000 LoC changed since last review) AND the caller has Agent/Task-tool dispatch capability (.e., top-level orchestrator or main Claude session). Trigger on phrases like "code-review backlog is big", "no review for weeks", "many commits without review", "review of 100+ commits", "how do I review these 200 commits", "chunk review", "parallel code-reviewer subagents". Do NOT load for single-PR review (use superpowers:requesting-code-review directly), for <30-commit backlog (single subagent suffices), for non-Git codebases (chunking-by-SHA-range presumes git workflow), or when running as a subagent yourself without Agent-tool access (the chunk-dispatch pattern is then unexecutable — use risk-based-triage instead, documented in the skill's Fallback section). Complements (sub-of) `superpowers:requesting-code-review` and `superpowers:dispatching-parallel-agents`.
Code review findings als red tests
Ed3Design/ed3design-skill-bundles/code-quality/skills/code-review-findings-als-red-tests Skill
0★ repoUse when a code-review subagent (e.g. feature-dev:code-reviewer) has returned Critical/Important findings on a feature commit and you're about to write fixes. Standard default is "direct-fix, no new test". This skill says: each finding is itself a test case — write a RED test that shows the bug behavior, then GREEN fix. Prevents regressions + documents the bug pattern for future reviews. Trigger on phrases like "fix code-review findings", "implement Critical fix", "review subagent found C1/I1/I2", "convert reviewer output to fix", "close review cycle". Do NOT load for trivial style findings (typo, unused-import), for findings without a reproducible code path (perf hints, future-proofing), for code where TDD was consciously skipped (user-confirmed), or when the subagent had NO findings.
Commit message honesty precheck
Ed3Design/ed3design-skill-bundles/code-quality/skills/commit-message-honesty-precheck Skill
0★ repoUse before writing or finalizing a git commit message, especially when the subject line uses words like "NO-OP fix", "no behavior change", "refactor only", "trivial", "cleanup", "minor", "simple bump", "doc-only" — these phrases are often LIES in disguise that hide real semantic changes. Trigger on phrases like "small fix", "just a refactor", "NO-OP", "behavior unchanged", "trivial change", "harmless cleanup", "just bumping the default", or before any commit that touches a default value, signature, env-var fallback, or filter parameter. Do NOT load for genuinely cosmetic changes (typos, comments-only, whitespace), for first-line spelling, or for non-git contexts (slack messages, jira tickets).