agentsclimarketplace

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

  • Gh pr

    chienchuanw/chuan-skills/plugins/git/skills/gh-pr Skill

    no license0★ repo

    Push a development branch and create or update a pull request via gh CLI using type-specific templates. Optionally updates README, docs, and project tracking before opening the PR. Use after finishing development, when ready to open a PR, submit work for review, or push changes.

  • Pre push test

    chienchuanw/chuan-skills/plugins/git/skills/pre-push-test Skill

    no license0★ repo

    Use when the user wants to prevent broken code from being pushed, block git push when tests fail, run tests before push, set up push protection, or add a safety net against pushing untested code. Covers any request to gate pushes on test results — whether they mention "hook", "pre-push", or just describe wanting to stop bad code from reaching the remote. Also handles auto-fix/auto-repair of test failures before push. Trigger for: "block push if tests fail", "run tests before pushing", "protect branch from broken pushes", "stop pushing untested code", "catch failures before they reach remote", or any test-gated push workflow regardless of language (npm, pnpm, cargo, go, pytest, etc).

  • Portfolio review

    chienchuanw/chuan-skills/plugins/portfolio/skills/portfolio-review Skill

    no license0★ repo

    Read-only thinking check for the Obsidian portfolio at Personal/Finance/Portfolio/. Reads every open ticker file, compares the original opening thesis against later thesis-log entries and the most recent price action, and surfaces drift (you bought for reason A but are now justifying with reason B), anchoring (you keep restating the same thesis without re-examining it), and stale theses (no new entry in months despite price/news moves). Produces a markdown review the user can paste into their Journal — it challenges your *reasoning* and never tells you to buy or sell. Use this whenever the user wants to sanity-test their investing *thinking* rather than get moves — "is my thesis still right", "challenge my thinking on X", "am I anchoring", "are my reasons still valid", "thesis check", "portfolio retrospective", "重新檢視持股的理由", "我的持股邏輯還成立嗎". For actionable buy/sell/rebalance suggestions and directional calls — "what should I do", "any moves to make", "should I trim X", "投資建議" — use portfolio-advisor instead, which chains this review as its first phase and then escalates to the calls this skill refuses to make. Pair skill of portfolio-update — portfolio-update WRITES state from screenshots, this skill READS state and challenges thinking. Read-only: never edits portfolio files.

  • Gotcha capture

    chienchuanw/chuan-skills/plugins/skill-optimize/skills/gotcha-capture Skill

    no license0★ repo

    Use when the user encountered a problem, bug, edge case, or frustration with an existing skill and wants to save that knowledge into the skill's file. Trigger phrases include "add a gotcha", "document this pitfall", "capture lessons learned", "record edge cases", "pull out what went wrong", but also trigger when the user simply describes something that went wrong or was painful with a named skill (like readme, commit-msg, seo-meta, branch-report, gh-pr) and asks to update it, save it, or make sure it doesn't happen again. This skill mines conversation history for failure knowledge and writes it into the skill's Gotchas section. Do NOT use for creating new skills, adding features to skills, or changing how skills behave — only for documenting what went wrong.

  • Chengyu review old know new

    evnchn-agentic/chengyu-skills/chengyu-review-old-know-new Skill

    0★ repo

    Use when bringing forward priors from PAST sessions / cross-agent history before doing new work — NOT about re-reading the current context window (which is trivially in-context). Specifically triggered when auto-compact is lossy, when cross-session reference is needed, when the same topic has been treated before, when the original work was done in a different agent's session, or when you need to locate/date a past session JSONL or answer "have we done this before / what did we decide". The discipline is targeted historical retrieval via grep + read of session JSONLs (incl. the mtime/substring/leading-dash traps that make naive search silently return nothing), then promotion to memory if the prior deserves persistence.

  • Chengyu seeing is believing

    evnchn-agentic/chengyu-skills/chengyu-seeing-is-believing Skill

    0★ repo

    Use when about to claim work is done / verified / working — empirical verification is required, not model-based confidence. Run the code, observe the output, check the file exists, verify the page renders, read the actual stack trace. Don't trust your model of the system; trust the system. Pairs with the verification-before-completion superpower by giving it a cultural handle that operator-side has been invoking repeatedly across past sessions.

  • First touch review

    9uiLe/plugins/plugins/agent-ops/skills/first-touch-review Skill

    0★ repo

    Run fixed-rubric first-touch UX review loops on a real device or simulator, complementing static code/design review with evidence from actual operation. Japanese triggers: 「初見ユーザーとして評価して」「実機で触って評価して」「シミュレータで操作して評価」「UX を初見で評価」.

  • Quality review

    9uiLe/plugins/plugins/quality-architect/skills/quality-review Skill

    0★ repo

    Review EXISTING code, diffs, PRs, and repositories against the ISO/IEC 25010:2023 product quality model. Produces severity-rated findings, design-vs-implementation labels, deterministic static-analysis scorecards, and cited recommendations. Always attempts deterministic measurement before LLM judgement. Existing-code design validity reviews belong here; new or replacement architecture design belongs in `quality-architecture`. Japanese triggers: 「品質観点でレビューして」「25010 でレビュー」「実装をレビューして」「PR を品質特性で見て」「指摘して」「この設計は妥当か」.

  • Decision council

    9uiLe/plugins/plugins/second-opinion/skills/decision-council Skill

    0★ repo

    Make and document a consequential decision with evidence-gated, independent advisor positions, structured challenge, explicit dissent, and accountable synthesis. Use when choosing among approaches, asking multiple models such as Fable and Codex to debate, selecting a main agent or orchestration roles, or when the user asks why an option was chosen, what concerns remain, or whether information is sufficient. Do not use for a lightweight review with no decision; use second-opinion instead.

  • Second opinion

    9uiLe/plugins/plugins/second-opinion/skills/second-opinion Skill

    0★ repo

    Summon an external second-opinion reviewer over the CURRENT Claude Code or Codex session. Mechanically extract the conversation (no summarising by you) and send it unedited to Codex and/or Fable for a fixed five-section verdict: blind spots, convergence, ship/no-ship, decisive constraint, strongest counterargument. Backend, model, and effort are chosen at call time; if model/effort are unspecified, ask the user. Japanese triggers: 「セカンドオピニオンが欲しい」「アドバイザーに相談」「第三者の意見が欲しい」「Codex に見てもらって」「Fable に見てもらって」「この方針でいいか外部レビューして」「詰まったので別の視点が欲しい」.

  • Review

    vidit19sharma/neogents/skills/review Skill

    0★ repo

    Force an adversarial review — spawns smith in a fresh context to hunt flaws in the current changes. Use before merging, after finishing a feature, or when the user says "review this" or "check my work".

  • Devops safe ship

    YoniChechik/claude-code-config/skills/devops-safe-ship Skill

    no license0★ repo

    DevOps playbook for risky multi-step rollouts, migrations, and cutovers — incremental single-purpose PRs, inert-but-present fallbacks for one-line rollback, self-validation in prod (Chrome walkthroughs, /healthz probes, expected-vs-observed GO/NO-GO tables, negative controls), human-gated irreversible steps, CI-guard discipline, and durable milestone memory. Invoke for any high-blast-radius change e.g. auth/issuer cutovers, signing-key revokes.

  • Quality

    YoniChechik/claude-code-config/skills/quality Skill

    no license0★ repo

    Run quality checks, fix code quality issues, check code style, format code, or perform lint checks

  • Review tests

    YoniChechik/claude-code-config/skills/review-tests Skill

    no license0★ repo

    Review and fix tests for quality issues

  • Review

    YoniChechik/claude-code-config/skills/review Skill

    no license0★ repo

    Run comprehensive code review on current branch changes

  • Grill design

    qduc/claude-skills/skills/grill-design Skill

    no license0★ repo

    Investigate and pressure-test the design of a non-trivial software feature before implementation. Use when requirements, architecture, integration points, edge cases, compatibility, migrations, rollout, or testing decisions remain uncertain, especially in an existing codebase.

  • Review

    jmfontaine/jmf-claude-plugins/plugins/project-qa/skills/review Skill

    0★ repo

    Review project files for consistency, completeness, and correctness. Use when asked to review, audit, check, or validate project configuration files (justfile, Makefile, CI workflows, pre-commit config, pyproject.toml, Dockerfile, docker-compose, etc.) or documentation files (README, CLAUDE.md, CONTRIBUTING, etc.). Supports both single-file review and cross-file consistency checks.

    The same file is in 2 other repositories

  • Git commit message

    senior-sigan/llm-skills/skills/git-commit-message Skill

    no license0★ repo

    Generate descriptive commit messages by analyzing git diffs. Use when the user asks for help writing commit messages or reviewing staged changes.

  • Pr describe

    senior-sigan/llm-skills/skills/pr-describe Skill

    no license0★ repo

    Generate PR title (conventional commits) and description by analyzing branch changes. Use when creating or describing a pull/merge request.

  • Apply review

    gagip/gagip-dev/plugins/common/skills/apply-review Skill

    no license0★ repo

    GitHub PR 리뷰 코멘트를 가져와 분석 보고서를 마크다운 파일로 저장한 뒤, 사용자 승인 후 코드 수정까지 수행. "PR 리뷰 반영해줘", "코드리뷰 코멘트 처리해줘", "리뷰 적용해", "리뷰 피드백 수정해줘", "review comment 처리", "리뷰 보고서 만들어줘" 같은 요청에 반드시 이 스킬을 사용할 것.

  • Commit

    gagip/gagip-dev/plugins/common/skills/commit Skill

    no license0★ repo

    변경사항을 분석하여 커밋 메시지를 작성하고 git 커밋을 수행하는 스킬. 사용자가 "커밋해줘", "변경사항 정리해줘", "커밋 메시지 만들어줘", "git commit" 등의 표현을 사용할 때 반드시 이 스킬을 사용할 것.

  • Android review

    gagip/gagip-dev/plugins/mobile/skills/android-review Skill

    no license0★ repo

    코드 리뷰를 수행한다. 사용자가 "리뷰해줘", "코드 봐줘", "PR 리뷰", "변경사항 확인", "코드 점검", "코드 검토" 등 코드 품질·개선 관련 요청을 할 때 반드시 사용한다. 인자 없으면 git diff 기준, 인자 있으면 지정 파일/폴더 기준으로 가독성·정확성·보안·아키텍처·테스트를 체크한다.

  • Review quality

    gagip/gagip-dev/plugins/mobile/skills/review-quality Skill

    no license0★ repo

    모바일 앱 코드베이스를 플랫폼 공식 품질 기준(Android Core App Quality + Apple App Review Guidelines/HIG)에 비춰 정적 검토하고, 위반·미흡 항목을 근거와 함께 리포트한다. 사용자가 "앱 품질 검토", "출시 전 점검", "스토어 심사 전 확인", "Play/App Store 리젝 위험 봐줘", "플랫폼 가이드라인 위반 확인", "Core App Quality", "App Review Guidelines", "권한/네트워크 보안 점검", "접근성/다크모드 점검", "이 앱 심사 통과될까", "Wear OS 품질", "워치 앱 출시 점검", "Wear App Quality" 같은 요청을 하면 반드시 사용한다. 특정 코드 로직 버그를 찾는 일반 코드 리뷰가 아니라, "플랫폼이 요구하는 출시 품질 기준" 관점의 점검이다. Android·iOS 네이티브와 React Native·Flutter·Tauri·Capacitor 하이브리드, 그리고 Wear OS(워치) 앱을 모두 다룬다.

  • History manager

    RainLib/full-stack-skill/.claude/skills/history-manager Skill

    0★ repo

    Manage workflow iteration state and history. Use when any phase skill needs to check progress, create iterations, update phase status, or verify document existence. Invoked by dev-workflow and all phase skills.

  • Self verification

    RainLib/full-stack-skill/.claude/skills/self-verification Skill

    0★ repo

    代码开发完成后运行全量测试、检查集成、代码质量自检,产出验证报告。在代码开发之后使用,为全栈工作流最后一步。

  • Technical review

    RainLib/full-stack-skill/.claude/skills/technical-review Skill

    0★ repo

    评审需求与技术选型的架构可行性,定义 API 契约、数据模型、识别风险项。在技术选型之后、程序设计之前使用。

  • Architecture impact review

    andresnator/agents-orchestrator/skills/architecture-impact-review Skill

    0★ repo

    Trigger: architecture impact review, layer boundaries. Decide whether legacy risk is local or architectural.

  • Fable planning

    andresnator/agents-orchestrator/skills/fable-planning Skill

    0★ repo

    Trigger: deep plan, plan like Fable, planificar a fondo, fable plan. Fable-5-style planning: evidence before opinion, outcome-first selectivity, minimal calibrated questions, explicit edge-case validation, end-to-end verification plan.

  • Grill

    andresnator/agents-orchestrator/skills/grill Skill

    0★ repo

    Trigger: grill me, grill with docs, grill me sdd, grill code, grill docs open, entrevistame, entrevistame sdd. Router for relentless interview modes: plain, docs, or SDD planning.

  • Input validation preconditions

    andresnator/agents-orchestrator/skills/input-validation-preconditions Skill

    0★ repo

    Trigger: input validation, preconditions, boundary validation. Detect missing or duplicated preconditions.

  • Skills

    ijufumi/claude-skills/skills/git-sync-cleanup/skills Skill

    no license0★ repo

    ローカルリポジトリを最新化し、不要になったローカルブランチを掃除するスキル。ベースブランチ(main / master を自動検出)に切り替え、`git pull --prune` で origin から最新化し、リモート側ですでに削除されているローカルブランチ(upstream が `gone` のもの)を一括削除する。実行前に未コミット変更を検知して中断し、削除対象は一覧提示してユーザーから一括承認を取り、未マージのブランチは安全のため削除せずリストで報告する。リモートは origin 固定。ユーザーが「ローカル最新化」「コードを最新化」「main を pull」「master 更新」「最新の main に追従」「ブランチ整理」「ブランチ掃除」「不要ブランチ削除」「リモートで削除されたブランチ」「マージ済みブランチを消したい」「git pull --prune」「git fetch --prune」「prune」「gone ブランチ」「朝イチの同期」「作業開始時の同期」「pull して掃除」「sync」「cleanup branches」「prune local branches」などに言及した場合にこのスキルを使うこと。「pull」「最新化」「同期」「ブランチ削除」だけでも、ローカル作業の起点としてこのスキルが適切であれば積極的に使ってよい。

  • Skills

    ijufumi/claude-skills/skills/project-review/skills Skill

    no license0★ repo

    リポジトリ全体(または指定スコープ)に対するプロジェクトレビューを行うスキル。差分ベースの code-review とは異なり、PR や git diff ではなく**現在のコードベース全体**を対象とし、ファイル/モジュール単位での指摘・アーキテクチャレベルの所見・横断的な傾向分析までを行う。**既定の通常モード(normal)では 1 つの subagent A_review が 11 観点すべてを横断的にレビューし、1 つの subagent C_review がそのメタレビュー(検証)を行う**。ユーザーが「詳細に」「詳しく」「観点別に」「観点ごとに」「detailed」「徹底的に」「thoroughly」などのキーワードを依頼文に含めた場合のみ、**詳細モード(detailed)**に切り替わり、11 観点(コード正確性 / プロジェクト規約準拠 / パフォーマンス / テストカバレッジ / セキュリティ / エラーハンドリング / 可読性・保守性 / シンプル化 / アーキテクチャ・設計 / リポジトリ共通観点 + ユーザー指定の重点観点)に分割し、**1 観点 1 subagent で並列レビュー / 観点ごとに別 subagent で評価**する。実行前に必ずユーザーへ3点を確認する — ①**レビュー範囲**(リポジトリ全体 / ユーザー指定のディレクトリ・glob)、②**動作確認**(テスト/lint/型チェック/ビルド等の実行検証を実施するか、静的レビューのみに留めるか)、③**出力先**(コンソール表示 / Markdown レポートファイル保存 / 両方)。指摘は 🔴 MUST / 🟡 SHOULD / 🟢 NICE TO HAVE の3段階で分類する。動作確認は独立した subagent B として並列に走らせる。リポジトリ共通のレビュー観点(docs/REVIEW.md)と、ユーザーからの重点観点も独立した観点として扱う。GitHub への投稿は本スキルの範囲外(PR レビュー用途は code-review スキルを使用)。ユーザーが「プロジェクトレビュー」「project review」「リポジトリレビュー」「コードベースレビュー」「全体レビュー」「リポジトリ全体をレビュー」「プロジェクト全体をレビュー」「コードベース全体をレビュー」「コードベースを評価」「リポジトリの品質チェック」「健全性チェック」「コードベース監査」「コード監査」「テックデット棚卸し」「アーキテクチャレビュー」「設計レビュー」「リポジトリの棚卸し」などに言及した場合にこのスキルを使うこと。「PR」「差分」「diff」「ブランチの変更」が含まれる場合は差分ベースの code-review スキルを優先する。

  • Critique

    Moliboy5000/.claude/plugins/cache/impeccable/impeccable/1.3.0/.claude/skills/critique Skill

    no license0★ repo

    Evaluate design effectiveness from a UX perspective. Assesses visual hierarchy, information architecture, emotional resonance, and overall design quality with actionable feedback.

  • Critique

    Moliboy5000/.claude/plugins/cache/impeccable/impeccable/1.3.0/source/skills/critique Skill

    no license0★ repo

    Evaluate design effectiveness from a UX perspective. Assesses visual hierarchy, information architecture, emotional resonance, and overall design quality with actionable feedback.

  • Flow deliver

    Moliboy5000/.claude/plugins/cache/nyldn-plugins/octo/9.30.0/skills/flow-deliver Skill

    no license0★ repo

    Multi-AI validation, scoring, and review using Codex and Gemini CLIs (Double Diamond Deliver phase). Use when: AUTOMATICALLY ACTIVATE when user requests validation, scoring, or review:. "review X" or "validate Y" or "test Z". "score this", "quality check", "validate before shipping"

  • Skill code review

    Moliboy5000/.claude/plugins/cache/nyldn-plugins/octo/9.30.0/skills/skill-code-review Skill

    no license0★ repo

    Expert multi-AI code review with quality and security analysis

  • Skill finish branch

    Moliboy5000/.claude/plugins/cache/nyldn-plugins/octo/9.30.0/skills/skill-finish-branch Skill

    no license0★ repo

    Wrap up a branch — run tests, create PR, merge or discard — use when implementation is done. Use when: AUTOMATICALLY ACTIVATE when user requests task completion with git operations:. "commit and push" or "git commit and push". "complete all tasks and commit and push"

  • Skill staged review

    Moliboy5000/.claude/plugins/cache/nyldn-plugins/octo/9.30.0/skills/skill-staged-review Skill

    no license0★ repo

    Review code in two passes: spec compliance then quality — use for thorough PR or feature reviews. Use when: Use when completing a major feature, preparing a PR, or when user says. "staged review", "full review", "review against spec", or "two-stage review".

  • Skill verify

    Moliboy5000/.claude/plugins/cache/nyldn-plugins/octo/9.30.0/skills/skill-verify Skill

    no license0★ repo

    Verify claims with actual evidence before declaring success — use to prevent false completion. Use when: Use when about to claim work is complete, fixed, or passing.. Auto-invoke before: commits, PRs, task completion, moving to next task.. ALWAYS use before expressing satisfaction ("Done!", "Fixed!", "All passing!").

  • Design review

    Moliboy5000/.claude/skills/design-review Skill

    no license0★ repo

    Claude dotfile containing skills, agents etc I use.

  • Devex review

    Moliboy5000/.claude/skills/devex-review Skill

    no license0★ repo

    Claude dotfile containing skills, agents etc I use.

  • Guard

    Moliboy5000/.claude/skills/guard Skill

    no license0★ repo

    Claude dotfile containing skills, agents etc I use.

  • Merge conflict resolver

    Moliboy5000/.claude/skills/merge-conflict-resolver Skill

    no license0★ repo

    Resolve git merge, rebase, and cherry-pick conflicts interactively. Use this skill whenever the user mentions merge conflicts, conflict markers, failed merges, rebase conflicts, or asks to resolve conflicts in their repository. Also trigger when the user says things like "fix my merge", "I have conflicts", "help me rebase", or "resolve the conflicts". Even if they don't say "merge conflict" explicitly, if context suggests they're dealing with conflicting changes in git, use this skill.

  • Plan ceo review

    Moliboy5000/.claude/skills/plan-ceo-review Skill

    no license0★ repo

    Claude dotfile containing skills, agents etc I use.

  • Plan design review

    Moliboy5000/.claude/skills/plan-design-review Skill

    no license0★ repo

    Claude dotfile containing skills, agents etc I use.

  • Plan devex review

    Moliboy5000/.claude/skills/plan-devex-review Skill

    no license0★ repo

    Claude dotfile containing skills, agents etc I use.

  • Plan eng review

    Moliboy5000/.claude/skills/plan-eng-review Skill

    no license0★ repo

    Claude dotfile containing skills, agents etc I use.

  • Review

    Moliboy5000/.claude/skills/review Skill

    no license0★ repo

    Claude dotfile containing skills, agents etc I use.

  • Commit

    Stoica-Mihai/claude-skills/plugins/commit/skills/commit Skill

    0★ repo

    Handles git commits with concise, list-style messages and no co-author trailers. Use this skill whenever the user asks to commit, says "commit", "/commit", "commit this", "save my changes", or any variation of requesting a git commit. This skill MUST be used for every commit operation — it overrides the default commit behavior.

  • Dry principle

    Stoica-Mihai/claude-skills/plugins/engineering-principles/skills/dry-principle Skill

    0★ repo

    Enforces DRY (Don't Repeat Yourself) thinking across all programming tasks — writing new code, reviewing code, fixing bugs, refactoring, editing config, schemas, tests, and documentation. Use this skill for ANY coding or programming-related task: feature implementation, bug fixes, code review, refactoring, writing tests, editing configuration, database schema changes, build system modifications, or documentation updates. This skill should trigger whenever Claude is about to write, modify, or review code or code-adjacent files. Even if the user doesn't mention "DRY" or "duplication", this skill applies to all software engineering work.

  • Fact check

    Stoica-Mihai/claude-skills/plugins/engineering-principles/skills/fact-check Skill

    0★ repo

    Enforces evidence-based reasoning for any task that involves making factual claims about a codebase — debugging, bug fixing, code investigation, code modification, AND explaining what code does, answering "where is X" / "what does Y do" / "is this safe", or summarizing behavior. Use this skill whenever the user asks to fix a bug, investigate an issue, modify existing code, trace a problem, debug behavior, refactor, change, update code, or explain/locate/audit any part of the codebase. This skill ensures Claude gathers real evidence from source code, docs, git history, and runtime behavior before making any claim — never guessing, never paraphrasing comments as fact, never relying on training memory. Even if the task seems straightforward, use this skill to guarantee that every claim and recommendation is grounded in verified facts.

  • Smoke test

    Stoica-Mihai/claude-skills/plugins/quality-assurance/skills/smoke-test Skill

    0★ repo

    Designs and scaffolds smoke-test suites (build-verification tests) for web services, APIs, and software systems. Use this skill whenever the user asks to "add smoke tests", "set up build verification", "write a fast test that proves the deploy isn't broken", "what should my smoke tests cover", "wire smoke tests into CI", or anything else about catching catastrophic failures before downstream testing. Also trigger on "BVT / BAT", "intake testing", "sanity vs smoke", "build acceptance". The skill detects the project's stack (Python/FastAPI, Node/Express, Go net/http, Rust/axum, etc.), proposes a focused 5-10 check suite following the API-first, <2-minute, idempotent discipline, and emits a runnable scaffold in the right framework — with `TODO` markers for environment-specific values like base URL and auth tokens. Even if the user has not explicitly said "smoke", trigger when they ask for a "fast CI gate", "deploy-readiness check", "is-the-build-up test", "happy-path coverage", or describe symptoms like "we keep running a 20-minute regression suite against broken builds".

  • Create linear idiom issues

    pambrose/pambrose-claude-plugins/linear-integration/skills/create-linear-idiom-issues Skill

    no license0★ repo

    This skill should be used when the user asks to "check for non-idiomatic code", "review idioms", "find non-idiomatic patterns", "check idiomatic usage", "review code style", "find Java-style Kotlin", "find unidiomatic code", or wants a systematic codebase review focused on non-idiomatic language usage that files findings as Linear issues.

  • Document

    sujeet-pro/agents-devkit/plugins/adk/skills/document Skill

    0★ repo

    Draft a technical document from an intent or a source. Triggers on "write the runbook / ADR / RCA / PR description / commit message / changelog / diagram / README / migration guide / API reference / experiment report / incident summary / onboarding doc / design doc". Markdown-first: produces a draft in a local file and NEVER publishes (no Confluence / Jira / Slack / GitHub posting — that is a separate concern). Reader-first voice: leads with the reader's question, cites every non-trivial claim to a repo path or quoted source, caps external quotes at 15 words, cuts filler. Audience-tuned (engineer / pm / exec / mixed) — the voice does not mix. GitHub context (PR / issue) is read via the gh CLI.

  • Implement

    sujeet-pro/agents-devkit/plugins/adk/skills/implement Skill

    0★ repo

    Implement, build, ship, or wire up a change in an existing repo. Polymorphic on input: a Jira URL/KEY-123, a GitHub issue URL or #N (fetched via the gh CLI), a Confluence/Slack link (fetched via MCP), or freeform "build the X" prose. Plans before it acts: gathers context, presents 2-4 approaches with trade-offs, confirms, then writes the smallest correct change matching repo conventions, with tests for new behavior. Validates with the repo's own typecheck/lint/narrow tests at every checkpoint. Git via git directly (feature branch, never protected, never force, never merge); GitHub via the gh CLI only (PRs via gh pr create); clones SSH-only.

  • Pr review

    sujeet-pro/agents-devkit/plugins/adk/skills/pr-review Skill

    0★ repo

    HEAVYWEIGHT, GitHub-only deep PR review. Takes a GitHub PR URL, prepares a read-only git worktree at the PR head (SSH clone only), and reviews the diff with full cross-file context via Read/Grep/Glob. Fans out one agent per dimension (correctness, security, tests, performance, api, docs, observability, concurrency, feature-flow) through the Workflow tool, then adversarially verifies every finding before it survives. Traces feature flags / experiments via the statsig MCP and pulls linked Jira/Confluence via the atlassian MCP. Posts inline review comments, a review summary, and appreciations through the gh CLI after confirmation. NEVER merges, never force-pushes, never edits the PR's code. For a quick read-only pass with no worktree, use /adk:review.

  • Review

    sujeet-pro/agents-devkit/plugins/adk/skills/review Skill

    0★ repo

    Review, audit, sanity-check, look-at any review-able target. Triggers on a GitHub PR URL (fetched via the gh CLI), a local path or "." (review the working tree), a markdown/doc file, or a comment-thread URL. Read-only by default; --fix applies accepted findings locally and pushes to the PR branch after confirmation (never force, never merge, never to a protected branch). Produces severity-tiered findings (blocker / critical / should / may / nit) with path:line and <=15-word verbatim evidence quotes. Six dimensions: correctness, tests, security, performance, readability, consistency. For a deep PR review with cross-file code-context retrieval, use /adk:pr-review.

  • Commit

    ForeverAProgrammer/claude-devflow/skills/commit Skill

    0★ repo

    Generate a git commit message and apply it immediately.

  • Create branch

    ForeverAProgrammer/claude-devflow/skills/create-branch Skill

    0★ repo

    Create and checkout a git branch named to match the current uncommitted changes.

  • Create issue branch github

    ForeverAProgrammer/claude-devflow/skills/create-issue-branch-github Skill

    0★ repo

    Create a branch linked to a GitHub issue and check it out locally.