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
Cross repo state inspection before commit
Ed3Design/ed3design-skill-bundles/code-quality/skills/cross-repo-state-inspection-before-commit Skill
0★ repoUse when the user issues a vague git instruction like "commit hub", "push that to the repo", "should go in git", "put that under version control", "stage the changes", "deploy that to GitHub" — without specifying scope (which files, which repo, which branch, push-or-just-commit, new-repo-or-existing). Jumping to action without verifying repo state risks picking the wrong scope: committing everything when only one subdir was meant, creating a new repo when one already exists, pushing modifications meant to be deferred. The agent must run `git status` + `git log origin/main..HEAD` + `git remote -v` before acting. Trigger phrases like "commit X", "push that", "into the repo", "into git", "stage that", "bring to GitHub", "new commit needed". Do NOT load when the user has given fully-specified instructions, for clearly-single-project repos, when the working tree is clean, or for fresh `git init` workflows.
Frontend ui self verify before user demo
Ed3Design/ed3design-skill-bundles/code-quality/skills/frontend-ui-self-verify-before-user-demo Skill
0★ repoUse when a frontend/UI change has been deployed (Vite/React/Vue, FastAPI Jinja, Electron, any browser-rendered surface) AND you are about to tell the user "refresh the browser", "check the app", "it's deployed" — STOP and self-verify first via Playwright/Chrome-DevTools-MCP: `browser_navigate` → `browser_snapshot` → `browser_take_screenshot` → `browser_evaluate` (DOM-query the element/text/style you changed). Especially required when (a) the surface has multiple modes (edit/preview/print, light/dark, mobile/desktop) and the user may view a different one than you changed, (b) the user already complained "nothing changed" — switch to verify-mode, do NOT add cache-busting theories, (c) you rebuilt >2 times without visual confirmation. Do NOT load for backend-only changes, for changes the user said to deploy without checking, or for features needing auth/state a headless browser cannot reach. Prevents burning debug-rounds on cache-theories when the change rendered in a mode the user wasn't looking at.
Ed3Design/ed3design-skill-bundles/code-quality/skills/pre-push-bypass-audit-trail Skill
0★ repoUse when a `git push` is blocked by a pre-push hook (freshness/lint/test/data-pipeline) AND the blocking condition is provably orthogonal to the commits being pushed (different subsystem, different pipeline, known pre-existing issue). Trigger on phrases like "pre-push hook blocks", "--no-verify with justification", "hook drift", "freshness check failed", "push aborted but code is OK", "should use --no-verify". Produces (1) orthogonality check before bypass, (2) audit-trail entry in the Daily Note with reason + blocking issue + backlog reference, (3) a separate backlog item for the blocking issue, (4) the `--no-verify` push only after these are done. Do NOT load when the hook block is directly related to the commits, when no Daily Note exists, when bypass has not been explicitly approved by the user, or when the blocking issue is critical/security.
Roadmap phase execution verify first
Ed3Design/ed3design-skill-bundles/planning-disciplines/skills/roadmap-phase-execution-verify-first Skill
0★ repoCaptures a pattern from a Phase-A execution session: when executing a roadmap phase from a knowledge vault that was designed days/weeks ago AND a previous session may have prepared draft artifacts (service maps, status files, execution guides), verify each item against current reality (DB queries, docker inspect, systemctl, filesystem) BEFORE applying changes. Find drift in pre-existing files; document discoveries that turn a routine item into an issue; never check a roadmap item as done just because the roadmap said so. Trigger on phrases like "implement Phase X", "execute roadmap item", "implement Phase A of … consolidation", "item Y from the roadmap", "execute the phase from the vault". Do NOT load for fresh roadmap design (use brain-dump-to-phased-roadmap), for executing a freshly-written task list with no vault history, or for non-vault roadmaps.
Strategic proposal vault persistence check
Ed3Design/ed3design-skill-bundles/planning-disciplines/skills/strategic-proposal-vault-persistence-check Skill
0★ repoUse at session-end (or mid-session when about to wrap a strategic discussion) when the session has produced structured strategic content — multi-row tables (priority/effort/value matrices, tool inventories), JSON/data schemas (API responses, config-shapes), multi-step recommendations ("Stage 1:..., Stage 2:..."), 5-category frameworks, or anything that took >10min to structure. Risk: this content lives in the conversation buffer, NOT in persistent knowledge — when context expires the entire structuring effort is lost. Trigger phrases like "session wrap-up", "end of day", "is that in the vault?", "did Claude capture that?", "summarize for later", "remember", "wrap up". Do NOT load for sessions producing only ephemeral content (chat-only, single-file-edit, debug-trace), mid-task while strategic discussion is still active, or when post-session-skill-review is already covering the same scan.
Enum known values via insert grep
Ed3Design/ed3design-skill-bundles/schema-discipline/skills/enum-known-values-via-insert-grep Skill
0★ repoUse BEFORE writing or editing a Python-side constants/enum/validator-set that should match a DB-column's real value-space — `_KNOWN_X = {...}`, `VALID_<DIMENSION> = frozenset(...)`, `class XStatus(str, Enum)`, `pydantic.Field(..., regex='^(a|b|c)$')`, in-app filter-allowlists. STOP and grep ALL `INSERT INTO <table>` + ALL `<table>.<column> = ...` setter-lines + ALL UPDATE-statements that touch the column BEFORE deciding the value-set in Python. Trigger when phrases like "maintain the _KNOWN_SOURCES list", "define valid_phase_set", "build pydantic validator for column", "filter-allowlist for UI", "value X is silently accepted", "value Y is wrongly rejected" appear. Method: grep INSERT/setter/UPDATE → verify actual values → disambiguate table/column/mode-field. Do NOT load for greenfield-tables (no INSERTs exist yet), typed-Postgres-ENUMs (`\d` shows the list), or purely internal Python constants without DB context.
Pre migration data verification
Ed3Design/ed3design-skill-bundles/schema-discipline/skills/pre-migration-data-verification Skill
0★ repoUse before adding a constraint to an existing populated table (CREATE UNIQUE INDEX, ALTER ... ADD CHECK, ALTER ... NOT NULL, ADD FOREIGN KEY): query the existing data for would-be violations FIRST and write a cleanup-statement (UPDATE/DELETE) into the same migration block BEFORE the constraint-create. Skipping this → silent migration failure ("duplicate key" / "check violation" / "null value not allowed"), the entire migration rolls back and no error surfaces until the next deploy. Trigger on phrases like "add unique constraint to existing table", "CREATE UNIQUE INDEX migration", "ALTER TABLE ADD CHECK", "make column NOT NULL on existing data", "add foreign key to populated table", "migration silently failed". Do NOT load for first-time CREATE TABLE migrations (no existing data to violate), for column ADD migrations without constraints (defaults handle the gap), or for development DBs that can be wiped. PostgreSQL-focused; the pattern transfers to MySQL/SQLite/Oracle.
Production seed vs demo seed split
Ed3Design/ed3design-skill-bundles/schema-discipline/skills/production-seed-vs-demo-seed-split Skill
0★ repoUse when about to push a Code-Repo with seed/fixture-data containing real personal/customer/tenant data to a Remote (GitHub, GitLab) — even private. Triggers on phrases like "push repo for the first time", "create new private repo", "set up GitHub remote for <project>", "commit seed.py", "tenant data in code", "customer data in test data", "GDPR test data". Do NOT load for repos without personal data (pure tech repos like ESP32-firmware), for already-pushed repos (then it's git-filter-repo territory, different skill), or for ephemeral test-data without real-person reference. Encodes a property-management push-session pattern: 26 commits contained real tenant data + a .env demo credential; without the pre-push scan everything would have landed on GitHub.
Read only sql via regex validator
Ed3Design/ed3design-skill-bundles/schema-discipline/skills/read-only-sql-via-regex-validator Skill
0★ repoUse when exposing a read-only SQL endpoint to a less-trusted caller (browser via HTTP, MCP-Tool to LLM, internal dashboard with copy-paste query box, future API consumer). The endpoint accepts a SQL string and must reject all mutating statements before executing. Encodes the regex-based denylist+allowlist pattern: (1) strip SQL comments (`--` and `/* */`) BEFORE matching to prevent comment-smuggle bypass, (2) normalize whitespace + uppercase, (3) denylist regex for INSERT/UPDATE/DELETE/DROP/CREATE/ALTER/TRUNCATE/CALL, (4) allowlist regex for SELECT/WITH/EXPLAIN/ANALYZE start-tokens. Trigger phrases like "read-only SQL endpoint", "SELECT-only API", "SQL escape hatch", "DB query as HTTP endpoint", "MCP tool exposes raw SQL", "Postgres read API". Do NOT load when a real SQL parser is available (sqlparse, pglast), for write-allowed endpoints, stored-procedure invocation, or DBs where SELECT itself has side-effects.
Schema verify via information schema
Ed3Design/ed3design-skill-bundles/schema-discipline/skills/schema-verify-via-information-schema Skill
0★ repoUse when about to write or execute SQL against a PostgreSQL/TimescaleDB table whose exact schema (column names, types, deprecated-vs-active-variant) is not 100% certain from current context — especially in forensic debugging where multiple tables have similar names (`ohlcv_1d` vs `ohlcv_1d_hist`, `users` vs `users_v2`) or column names have shifted over time. Iron-Law: `SELECT column_name, data_type FROM information_schema.columns WHERE table_name=$1 ORDER BY ordinal_position` is the cheapest 30-second insurance against UndefinedColumn, false-positive-empty-result, and wrong-table-queried. Trigger on phrases like "writing a query against table X", "UndefinedColumn", "column X does not exist", "verify DB schema", "I think the column is called Y", "before any forensic query". Do NOT load for ORM-mediated queries (SQLAlchemy/Django catch drift at model-load), first-time-schema-design, catalog-queries on information_schema itself, or connection issues.
Ed3Design/ed3design-skill-bundles/skill-system-meta/skills/post-session-skill-review Skill
0★ repoUse at the end of any productive session, when writing a Daily Note, when the user says "wrap up" / "remember" / "session done", or proactively after observing repeated workflow patterns within a session. Trigger on phrases like "end of session", "let's wrap up", "end of day", "session wrap-up", "end of day review", "what did Claude learn today", "is there a skill in here?", "post-session review", "skill candidate review". Do NOT load mid-session for active task work, for single-task sessions with nothing repeatable, or when the user is in mid-debug (they need closure, not meta-reflection).
Subagent mode selection continuous vs review between
Ed3Design/ed3design-skill-bundles/skill-system-meta/skills/subagent-mode-selection-continuous-vs-review-between Skill
0★ repoUse when starting a multi-task workflow via `superpowers:subagent-driven-development` (SDD) and the user must choose between two execution modes: (A) Continuous Execution — all tasks dispatched back-to-back without checkpoint (SDD default), or (B) Review-Between-Tasks — user inspects each task's output before authorizing the next. Mechanical-implementation tasks (TDD with complete code in plan, deterministic outcomes) are continuous-safe. Judgment-heavy tasks (forensics, interpretation of numbers, strategic decisions, architectural pivots based on subagent output) require review-between because the next task's prompt depends on prior outcome interpretation. Trigger phrases like "should I run continuous or with pauses between tasks", "review between steps", "subagent-driven with checkpoints", "pause after each task". Do NOT load for single-task workflows, non-SDD workflows, purely mechanical TDD with complete code in plan, or the very first task of a workflow.
isanthoshgandhi/santhoshstack/skills/founder-check Skill
0★ repoPsychological state audit for founders. Surfaces fear-based avoidance, decision clarity, conviction level, and co-founder alignment — then flags where psychology is driving business decisions. Not therapy. A diagnostic with an output. Invoke when the user says "founder check", "how am I doing", "am I being rational", "I'm avoiding something", "something feels off", or /founder-check.
isanthoshgandhi/santhoshstack/skills/review Skill
0★ repoScan the current codebase or a specified file/diff for production bugs, logic errors, and risky code before it ships. Use when the user says "review", "check my code", "is this safe to ship", or "any bugs in this?". Produces a prioritised bug list with exact file:line references and optional auto-fixes.
isanthoshgandhi/santhoshstack/skills/runway-check Skill
0★ repoFinancial survival check for startups. Takes burn, revenue, and growth rate — outputs a single verdict: DEFAULT ALIVE or DEFAULT DEAD, runway in months, and the exact lever that flips the verdict. No estimates accepted. Forces honest numbers. Invoke when the user asks "can we survive?", "how long do we have?", "are we default alive?", "runway check", or /runway-check.
Timelnapp/skills/skills/consulting-os/consult-consistency-lint Skill
0★ repoTrigger on "consult lint", "consistency pass", "diagram vs narrative", "broken promises across docs", "cross-artifact check", "scope drift lint". Read-only diff across frame, arc, gates, acceptance, commercial, architecture artifact. NOT rewriting copy (use consult-package), NOT drawing diagrams (use arch-diagrams).
Timelnapp/skills/skills/consulting-os/consult-integrity Skill
0★ repoTrigger on "consult integrity", "integrity gate", "integrity check", "verify proposal claims", "hallucination check proposal", "cos integrity". Seven-mode failure checklist at gates 3.5 and 5.5. Uses timeln-quickly and timeln-shipped to verify quotes and proof claims. NOT lint (use consult-consistency-lint), NOT red team (use consult-red-team).
Timelnapp/skills/skills/thinking-os/timeln-warned Skill
0★ repoTrigger on "what bit me on X", "have I been burned by X", "gotchas with X", "what went wrong last time with X", "warn me about X", "any retros on X", "past incidents with X". Use when the user wants to sanity-check an approach against their own failure history. NOT for general pitfalls (ask Claude directly) or past decisions (use timeln-decided).
robmcquade/claude-skills/optimize-efficiency Skill
0★ repoReviews a target — code, a script, a file, a process, a workflow, a prompt, a config, or a plan — for efficiency across four axes (resource use, code, speed, token usage) and returns prioritized, location-anchored recommendations, each carrying a per-candidate no-regression verdict, never a rewrite. Every recommendation must clear a forced guardrail: it may not reduce effectiveness, correctness, safety, or security, and the reviewer must enumerate the inputs whose output changes so a behavior change cannot be waved through as "clean." Trade-offs are surfaced for the user to decide; regressions are listed as declined. Recommends; does not edit. User-invoked only. Use when something works but feels wasteful, slow, bloated, expensive, or token-hungry, or before scaling a process whose per-run cost multiplies. Defer to /simplify to apply code simplifications, /code-review for correctness and bugs, and /pressure-test for general (non-efficiency) critique. Skip when correctness is unverified — fix that first.
gmarmat/claude-pm-kit/.claude/skills/verify Skill
0★ repo[PM Twin] Sprint-boundary verification gate — runs every feature in the sprint through qa-feature, runs the sprint's end-to-end journey flows, aggregates evidence, and gates the next sprint.
ClydeShen/harness-skill/skills/engineering/harness-verify-before-move Skill
0★ repoComprehensive pre-phase-transition sweep. Verifies phase exit criteria, syncs code status (git/lint/build/tests), audits and cleans design docs, cleans implementation-notes, prunes stale documents and memory, syncs GitHub project state (milestones, issues, status labels, board), updates README, writes a fresh memory snapshot, then outputs a clean project snapshot. Use when completing a phase, before advancing to next phase, or when user says "verify before move", "clean up before moving", "sync everything", "project snapshot", "clean slate", "phase transition". Typically consumes ~8K–15K tokens / <10% of a context window.
richardkmichael/claude-rodin/skills/review-pr-companion Skill
0★ repoPackage a code review as a review branch -- small, reviewer-shaped fix commits plus a REVIEW_HANDOFF.md written for the author's coding agent -- pushed for the author to integrate, with no pull request opened. Use when reviewing a pull request or a local branch and you want to hand back working commits and the reasoning behind them, not just inline comments.
paidaxing1234/latency-hunter-toolkit/skills/risk-config-lint Skill
0★ repo逐项审查交易系统的风控配置(risk_config.json/yaml)与 pre-trade 风控代码,定位"会让风控形同虚设、实盘爆仓"的工程缺陷,精确到 文件:行,按严重度输出"风控体检报告"。覆盖限额/回撤/杠杆/kill-switch/防胖手指/保证金/爆仓模拟/并发在途订单/单点故障/运维/密钥。触发词(中):风控审查、风控体检、风控配置、仓位上限、敞口上限、回撤止损、最大回撤、kill-switch、爆仓保护、强平、维持保证金、防胖手指、限额、在途订单。触发词(英):risk config、risk limit、risk audit、kill switch、kill-switch、fat finger、fat-finger、liquidation、margin、drawdown stop、position limit、in-flight order、risk-config-lint。不适用:策略 alpha/信号逻辑/调参寻优、行情数据源运维、回测可信度审查(用 backtest-guard)、连接器可靠性(用 connector-forge)、非风控的通用代码审查;也绝不承诺"绝对不爆仓/绝对安全"——本 skill 只做工程审查,不替代实盘风控演练。
Cadasto/go-coding-plugin/skills/go-lint-setup Skill
0★ repoScaffold the reference golangci-lint v2 config into a Go repo. This skill should be used when the user runs `/go-lint-setup` or asks to "set up", "scaffold", "add", or "bootstrap" golangci-lint or a `.golangci.yml` for a Go project — it writes the plugin's reference v2 config (modernize + stack linters) and will not overwrite an existing config unprompted. For understanding what the linters do or migrating a v1 config, use `go-linting`. Not for non-Go projects.
Cadasto/go-coding-plugin/skills/go-linting Skill
0★ repogolangci-lint v2 setup and adoption for Go. This skill should be used when the user configures, upgrades, or debugs Go linting — the `.golangci.yml` file, the golangci-lint v2 schema (the versioned config, the `linters.default` set, the separate formatters section), the `modernize` linter, or stack linters (errorlint, bodyclose, rowserrcheck, sqlclosecheck, noctx, contextcheck). Not for what individual idioms mean (use the `go-idioms` skill) or writing rules by hand.
TomerAberbach/claude-config/skills/git-split Skill
0★ repo🤖 My Claude Code config!
Reward seeking evidence review
jpoindexter/reward-seeking-safety-skills/skills/reward-seeking-evidence-review Skill
0★ repoReview evidence and calibrate claims about reward-seeking, reward hacking, grader targeting, evaluation gaming, or oversight-dependent behavior. Use before saying a model is safe, aligned, deceptive, reward-seeking, or improved; also use to distinguish executed behavior, code-path evidence, assumptions, and inconclusive null results.
tarangdeep-goel-by/claude-harness/claude/skills/review-merge Skill
0★ repoReview a PR AND merge it if it passes — find the current branch's PR, assess it against a quality rubric (correctness/tests/clean/style/scope), then `gh pr merge --squash --delete-branch`. Use when the user says /review-merge, 'review and merge this PR', 'ship this PR', or asks to merge a pull request. Distinct from /review (review-only, no merge) — review-merge is the merge moment. Use `--no-merge` for review-only.
ccronca/dotfiles/claude/skills/container-k8s-validator Skill
0★ repoValidates Containerfile COPY commands and Kubernetes manifest paths for consistency. Automatically runs when working with Containerfiles or K8s manifests to ensure source paths exist and destinations match between files.
ccronca/dotfiles/claude/skills/gemini-reviewer Skill
0★ repoUse Google Gemini for parallel code reviews, complex task validation, and second opinions. Runs Gemini CLI non-interactively to provide alternative perspectives on code quality, implementation decisions, and technical solutions.
ccronca/dotfiles/claude/skills/pragma-review Skill
0★ repoFetch and analyse Pragma's AI review for a GitLab MR. Summarises high-value findings and critiques the review quality. Usage: /pragma-review [mr_number]
chenwei791129/agent-skills/hermes-skills/superpowers/receiving-code-review Skill
0★ repoUse 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 implementation
chenwei791129/agent-skills/hermes-skills/superpowers/requesting-code-review Skill
0★ repoUse when completing tasks, implementing major features, or before merging/opening a PR/MR to verify work meets requirements
chenwei791129/agent-skills/hermes-skills/superpowers/using-git-worktrees Skill
0★ repoUse 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 fallback
Verification before completion
chenwei791129/agent-skills/hermes-skills/superpowers/verification-before-completion Skill
0★ repoUse 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 always
nishide-dev/claude-code-ml-research/skills/ml-format Skill
0★ repoFormat Python code with ruff formatter and optionally fix auto-fixable linting issues. Use when formatting code, preparing code for commit, or ensuring consistent code style across the project.
nishide-dev/claude-code-ml-research/skills/ml-lint Skill
0★ repoRun comprehensive code quality checks with ruff (format, lint) and ty (type checking). Use when checking code quality, fixing linting errors, or ensuring code follows best practices before commits or PRs.
mi4646/my-skills/skills/weekly-report Skill
no license0★ repo生成开发周报,根据 git commit 历史归纳本周/上周的工作成果。当用户提到"周报"、"工作总结"、"本周工作"、"上周工作"、"weekly report"、"写周报"、"生成本周总结"时触发此技能,即使用户没有明确说"周报"但表达了回顾近期工作的意图也应触发。
bogheorghiu/ex-cog-dev/research-toolkit/skills/iterative-verification Skill
0★ repoex-cog — externalized cognition. Four Claude Code plugins. research-toolkit — self-checking investigation & verification. makers-toolkit — build-discipline & prompt-design. vasana-system — cross-session pattern memory (forms, not just facts). security-toolkit — dangerous-action guardrails.
bogheorghiu/ex-cog-dev/security-toolkit/skills/pr-merge-guard Skill
0★ repoExplains and toggles the security-toolkit PR-merge guard — whether Claude is blocked from running `gh pr merge`. Fires in BOTH directions: when the user asks about it ("can you merge PRs?", "is there a setting that stops you merging?", "what's EXCOG_BLOCK_PR_MERGE?", "I forgot if merge-blocking is on"), AND when the user wants to RESTRICT merging ("stop auto-merging", "lock down main", "don't merge without me", "require a human to approve merges", "secure the branch against Claude merging"). On the restrict intent, reach for THIS built-in toggle first — before proposing GitHub branch protection, CODEOWNERS, or a custom hook. The guard is OFF by default. Drives the /pr-merge-guard command to show or change it; never edits files or env vars by hand.
Windows wsl security verification
bogheorghiu/ex-cog-dev/security-toolkit/skills/windows-wsl-security-verification Skill
0★ repoAm I compromised? After supply-chain news — a poisoned npm/PyPI package, a malicious VS Code extension, a backdoored dep that ran as you — this runs a guided IOC triage of a Windows + WSL2 dev box, then trims the attack surface the next compromise would use. WSL side: known-bad package versions, ld.so.preload, persistence (systemd/cron/autostart), planted SSH keys, shell-rc injection, executable .pth. Windows side: full AV scan (a third-party AV makes Defender passive — one scan, not two), second-opinion scan, Sysinternals Autoruns with VirusTotal, code --list-extensions, scheduled tasks, BYOVD-class drivers. Carries the discriminators that stop false alarms: filename IOCs via find not grep, VT named-family verdicts over aggregate labels, web-filter blocks vs infections. Fires on "am I hacked", "did I get owned", a named bad package or CVE in your dependency chain, or an AV detection you're unsure how to read. Windows/WSL-specific; verifies — does not harden. A clean result raises confidence, never proves.
bogheorghiu/ex-cog-dev/vasana-system/skills/check-assumptions Skill
0★ repoDid I actually verify this, or am I assuming? - Challenges technical decisions against behavioral patterns before proceeding. Be direct. Be challenging. Point out BS. Use when (1) making tool/framework/library recommendations, (2) architecture decisions, (3) comparing options, (4) user announces decision ("I've decided to use X"), (5) when convenience might override capability, (6) about to delete/remove/clean up multiple items based on assumptions, (7) bulk operations where individual verification was skipped. Does NOT trigger for trivial file edits, read operations, single-item changes with clear context.
bogheorghiu/ex-cog-dev/vasana-system/skills/test-pattern Skill
0★ repoTests whether patterns actually work — trigger, emerge, and produce value. Use when (1) user wants to verify a pattern before relying on it, (2) user runs /test-pattern command, (3) a pattern was just recorded and needs verification, or (4) user suspects a pattern isn't working as expected.
mjcreativelab/mjcreativelab-agent-prompts/skills/branch-visualize Skill
0★ repogit ブランチ単位で実装内容を可視化する。対象ブランチとマージ先ブランチの差分を分析し、 変更モジュール・ライブラリ増減・データ/ドメインモデルと直接の関連範囲を Mermaid / D2 / HTML(ノード数で自動選定、--format で明示指定可)の構成図として docs/branch-diagrams/ に出力する。--models でモデルの内部構造(クラス図/ER 図・ メンバー単位の差分マーカー付き)も描ける。対象はブランチ名のほか PR 番号(#123) でも指定できる。レビュー準備・作業内容の俯瞰に使う。 リポジトリ全体のアーキテクチャ図は作らない(変更箇所と直接関連範囲に限定)。 ユーザーが「ブランチを可視化して」「変更内容を図にして」「/branch-visualize」と言ったら起動する。
mjcreativelab/mjcreativelab-agent-prompts/skills/claude-code-update-review Skill
0★ repoClaude Code のバージョンアップ後に、最新の公式推奨手法と現在の構成 (settings.json / commands / rules / skills / CLAUDE.md 等)を照合して改善提案を行う。 「Claude Code の最新機能」「アップデートレビュー」「設定見直し」「ベストプラクティス確認」 「/claude-code-update-review」で起動。 単一スキルの品質改善 (skill-improver) や PR/差分レビュー (smart-review) とは別物。
mjcreativelab/mjcreativelab-agent-prompts/skills/code-reviewer-adversarial Skill
0★ repoBreaker(Claude)× Judge(Codex)の二者構造で、反例生成とテスト実行により「本当の欠陥」のみを抽出する敵対的コードレビュー。重要変更・最終ゲート・見逃したくない場面に使う。対象が PR なら確認ゲート経由で PR にレビューを投稿する。--claude-judge で Breaker・Judge を独立 Opus エージェント(コンテキスト隔離)にした claude-judge モードも使える(Codex 不在時は自動でこのモードにフォールバック・Workflow 前提)。通常のレビューは /code-reviewer を使うこと。
mjcreativelab/mjcreativelab-agent-prompts/skills/code-reviewer Skill
0★ repo実装されたコードを仕様整合・設計適合・可読性の観点でレビューする。PR / 変更差分のレビュー、コミット前のセルフレビューに使う。対象が PR なら確認ゲート経由で PR にレビューを投稿する。重要変更は Codex クロスチェックを推奨する。--isolated でコンテキスト隔離した Opus(effort max)エージェントによる単発レビューも可能(Workflow 前提・不能時はメインセッションに degrade)。
mjcreativelab/mjcreativelab-agent-prompts/skills/smart-commit Skill
0★ repo現在の git 差分を作業内容ごとに適切な単位で分割し、日本語 conventional commits でコミットする。 必要に応じて作業ブランチへの切替・新規作成も提案する(副作用あり)。 「コミットして」「差分をコミット」「変更をまとめて」「作業区切りたい」「一旦コミット」 「今の状態を保存」「ここまでの変更を記録」等で起動する。 コードレビュー(smart-review)や PR 作成(smart-pr)とは別物。
mjcreativelab/mjcreativelab-agent-prompts/skills/smart-git-sync Skill
0★ repoデフォルトブランチ(develop or main)にチェックアウトし、fetch/pull してマージ済みブランチを削除する。ユーザーが「同期して」「ブランチ整理」「/smart-git-sync」と言ったら起動する。
mjcreativelab/mjcreativelab-agent-prompts/skills/smart-pr Skill
0★ repo現在のブランチから GitHub Pull Request を作成または更新する。作業内容の自動要約・ラベル付与・関連 Issue 紐づけを行う。 必要に応じてデフォルトブランチを現在のブランチへマージしてから push する(副作用あり)。 「PR 作って」「プルリク作成」「PR 更新して」「/smart-pr」と言ったら起動する。 コミット(smart-commit)やレビュー(smart-review)とは別物。
mjcreativelab/mjcreativelab-agent-prompts/skills/smart-review-apply Skill
0★ repoPR レビューコメント・ローカルファイル・会話内のレビュー結果からフィードバックを読み取り、コードに反映する。ユーザーが「レビュー反映して」「フィードバック適用」「/smart-review-apply」と言ったら起動する。
mjcreativelab/mjcreativelab-agent-prompts/skills/smart-review Skill
0★ repoローカル変更をデフォルトブランチと比較してセルフレビューする。Issue 番号があれば要件適合もチェックする。ユーザーが「レビューして」「変更確認して」「/smart-review」「/smart-review #123」と言ったら起動する。
mjcreativelab/mjcreativelab-agent-prompts/skills/structure-visualize Skill
0★ repo指定された開発構造(インフラ構成、ER 図、コンポーネント設計、クラス構成、モジュール依存など)を 自己完結 HTML の構成図として docs/structure-diagrams/ に出力する。入力は会話で説明された設計・ 設計ドキュメント・リポジトリ内の実体(IaC / DB スキーマ / ソースコード)のいずれでもよい。 カテゴリ別配色とエリア枠(レイヤー・所属のグルーピング囲み)、ノード詳細パネル・ズーム/パン付き。 git ブランチ差分の可視化は /branch-visualize(本スキルは指定内容のスナップショット構造を描く)。 ユーザーが「構造を可視化して」「構成図にして」「/structure-visualize」と言ったら起動する。
JoseVelazcoH/python-skills/skills/python-code-review Skill
0★ repoTrigger: code review, review my changes, review this PR, check my code, pre-merge review, review diff. Review Python changes by the Review Pyramid, priority-first.
Fab2295/sap-skills/skills/sap-cap-test Skill
0★ repoTest-only skill for SAP CAP Node.js projects. Its sole purpose is to scaffold and run automated tests using `cds test` (Node.js test runner wrapper) and, when (and only when) explicitly requested by the user, produce coverage with `c8`. The skill writes test files under `test/` (or the project's existing test folder), executes them, and emits two possible report files at the project root: - `CAP-TEST-REPORT.md` — successful run summary (always) - `CAP-TEST-FAILURE.md` — failure report (only if a test fails) Use when the user asks to: - "cria testes", "gera teste para X", "scaffold tests" - "roda os testes", "executa cds test", "run tests" - "testa o serviço X", "cobertura c8" (coverage mode — must be explicit) Strict negatives — this skill NEVER: - edits production code (`srv/**`, `db/**`, `app/**`, `package.json`, `.cdsrc.json`, `mta.yaml`, `xs-security.json`) — it only writes inside the test folder - implements features, refactors, fixes bugs, changes business logic - runs `git add`, `git commit`, `git push`, `git merge`, `git rebase`, `git reset`, `git checkout` of source files, or any git mutation - installs dependencies on its own (it may *propose* `npm i -D c8` and wait for human approval; it does not run it unsolicited) - generates coverage unless the user explicitly opts in - touches the build output (`gen/`, `coverage/`, `dist/`) Scope: SAP CAP Node.js only. If the project is CAP Java or non-CAP, the skill refuses with an explicit message.
chienchuanw/chuan-skills/plugins/daily/skills/end-of-day Skill
no license0★ repoRun the user's bedtime wind-down — a pure recovery ritual to help sleep, reduce stress, and lightly consolidate the day's learning. It is deliberately the LAST thing of the day, run AFTER any accountability work: it NEVER touches todos, scoring, objectives, or 記帳 (those belong to daily-reviewer). A single-pass guided flow — the user answers one compact prompt block in one go — then the skill closes open mental loops, guides one short breath, appends one block to the month's wind-down log, and hands the user off the screen ("關螢幕、去睡") without asking anything further. Use whenever the user wants to wind down, decompress, or close the day for sleep — "end of day", "wind down", "我要睡了", "準備睡了", "睡前收尾", "收心", "幫我收個尾", "今天到此為止", "let me wind down before bed" — even if they don't name a skill. This is the bedtime bookend that comes AFTER daily-reviewer's retrospective: prefer daily-reviewer when the user wants to review, score, or 記帳 the day; prefer THIS skill when the user wants to relax, calm down, and go to sleep.
chienchuanw/chuan-skills/plugins/dev/skills/feature Skill
no license0★ repoEnd-to-end feature delivery workflow that orchestrates issue creation, branch setup, strict TDD implementation, PR opening, review-feedback handling, and session archival. Use when the user wants to ship a feature from start to finish with one command — e.g. "/feature 42", "/feature add RSS feed", "build this end-to-end", "deliver issue 17", "take this from issue to merged PR". Bundles the plan → TDD → PR → review-fix → archive arc that would otherwise require invoking 5+ separate skills.
chienchuanw/chuan-skills/plugins/git/skills/branch-report Skill
no license0★ repoGenerates a branch-vs-default comparison report with plain-language explanations and senior dev review. Use when the user asks to summarize, compare, review, or explain branch changes, or wants a PR summary before merging. Read-only: it produces a report to read — it does not push the branch or open/update the PR (for that use gh-pr).
chienchuanw/chuan-skills/plugins/git/skills/commit-msg Skill
no license0★ repoSuggests 3 commit messages based on git diff and project conventions. Use when the user asks for commit message ideas, wants help writing a commit message, or has changes ready to commit.
chienchuanw/chuan-skills/plugins/git/skills/gh-fix Skill
no license0★ repoRead unresolved review comments on a GitHub PR, deeply evaluate each suggestion, then either fix it (via gh-dev + simplify) and push the result, or push back via gh-comment with a reasoned reply. Use when the user wants to address PR feedback, respond to reviewer comments, fix code-review suggestions, action a review, or any phrasing like "address the comments on PR #N", "respond to the review", "fix what the reviewer asked for", "go through the PR feedback", or "handle the review comments".