Coding & Dev Tools
Coding agent skills teach AI agents repeatable engineering workflows: reviewing pull requests, writing tests, refactoring safely, and enforcing house style. Install one SKILL.md and your agent applies the same checklist every time, whether you use Claude Code, Cursor, Codex, or another agent.
-
yanqingcheng Skill Questions WillUse when a next move is visible but they haven’t locked a real commitment — soft yes, drift, or “I should…”.
-
yanqingcheng Skill Real Point Load Bearing HingeUse when they have reasons on one conclusion but it still won’t settle — arguing both ways, or “if I only knew ___.”
-
yanqingcheng Skill Coaching SessionDefault entry for a coaching chat — use when they bring a live hard problem and want thinking partnership, not advice or you doing the work.
-
yanqingcheng Skill Questions OptionsUse when the problem is clear enough that moves would help — and you’re tempted to offer a wider life menu instead of moves at this problem’s edge.
-
yanqingcheng Skill Questions RealityUse when claims about the situation are vague or untested, or they’ve skipped what’s already been tried.
-
yanqingcheng Skill Strategy CoachingUse when they’re debating plans or tactics while what they’re trying to achieve, what’s hard about it, and how they’d overcome that are still muddled.
-
yanqingcheng Skill Questions TransferUse when the working arc is done or paused and you’re about to end without capturing what landed or what they’ll carry.
-
yanqingcheng Skill Real Point FocusingUse when something clearly matters but they can’t find words — “I don’t know how to put it,” gestures, language keeps failing.
-
yanqingcheng Skill Coaching Working MapUse when more than one concern is in the air and you’re losing which is which — or they keep blending separate pulls into one blob.
-
yanqingcheng Skill Pivot The Coaching SubjectUse mid-flow when they realize “wait — that’s not the thing,” or energy leaves the subject you thought was locked.
-
yanqingcheng Skill Real Point Trust DanceUse when they flip between two pulls, cancel themselves, can’t speak as one voice, or each side dismisses the other.
-
yanqingcheng Skill Choose The Coaching SubjectUse at chat open when the first issue may not be what they actually want to work on — energy hasn’t locked yet.
-
yanqingcheng Skill Getting To The Real PointUse when you’re about to plan or decide but the real issue still feels slippery, foggy, or “not quite it.”
-
tsaru23 Bundle Webapp BlueprintWebアプリケーションをゼロから作るときの設計書(技術スタック選定・アーキテクチャ・ セキュリティ・プライバシー/計測・品質基準)を対話で作成する。「Webアプリを作りたい」 「技術選定して」「設計書を作って」「0から1で作りたい」「新規サービスを立ち上げたい」 「LPを作りたい」「3Dのサイトを作りたい」「セキュリティ要件を決めて」といった依頼で使用する。 既存プロジェクトへの機能追加・バグ修正・リファクタリングには使わない。
-
yanqingcheng Skill Leadership Lens RouterUse when the stuckness is about other people doing/following — not only their own solo decision.
-
yanqingcheng Skill Leadership Questions ExitUse when success stories all require them present — or they can’t step away without things falling over.
-
yanqingcheng Skill Leadership Questions GaugeUse when “leadership isn’t working” but they haven’t checked whether anything’s moving, how clear the destination is, or where the friction is.
-
yanqingcheng Skill Leadership Questions SituateUse when they leap to leadership actions without naming who the team is, what the goal is, or what the environment is doing.
-
yanqingcheng Skill Leadership Questions DeliveryUse when people agreed to work and it still isn’t happening — and the instinct is to push harder or take the wheel.
-
sivert-io Bundle SprakvaskNorsk språkvask etter Språkrådets normer og klarspråksråd. Bruk når teksten som skrives eller rettes er på norsk – bokmål eller nynorsk – i grensesnitt, e-post, dokumentasjon, vilkår, commit-meldinger eller kommentarer i kode. Fanger særlig feilene som oppstår når norsk skrives av noen som tenker på engelsk.
-
tsaru23 Skill Session Title Refreshセッション名を最新の作業内容に更新する。手動でも定期ループからでも使う
-
slayer1dev Skill Multi AI DelegationDecide qual IA executa cada tarefa quando há várias disponíveis (Claude, Codex/OpenAI, Antigravity/Gemini) e invoca cada uma corretamente via CLI headless. Use antes de tarefas criativas, de volume, de front-end ou de segurança, para não gastar o modelo caro no que o barato resolve.
-
slayer1dev Skill File Queue OrchestrationPadrão para rodar tarefas de IA em fila baseada em arquivos, com isolamento em sandbox, versionamento em git e revisão humana obrigatória antes de aplicar. Use ao construir automação que deixa um modelo escrever ou alterar arquivos sem supervisão em tempo real.
-
lovethread Bundle Codex Maintenance Suite接手和维护已有代码库,包括项目摸排、Bug 根因诊断、小范围二次开发、验证和代码审查;适用于需要尊重既有约定与未提交改动的维护任务。不用于从零搭建项目或未经请求的大规模重构。
-
lovethread Skill Code Review审查已有项目的代码改动、功能实现或修复结果;适用于查找会影响正确性、安全性、兼容性和可维护性的具体问题,并评估验证是否充分。
-
lovethread Skill Verification验证已有项目中的修复或功能调整是否达到目标;适用于选择测试层级、检查边界行为并准确报告已验证与未验证内容。
-
lovethread Skill Bug Diagnosis诊断已有项目中的错误、异常结果和回归问题;适用于根因尚不明确,需要结合复现、日志、调用链和相近实现缩小范围的任务。
-
lovethread Skill Project Intake只读摸排一个已有代码库,识别项目规则、结构、入口、已有改动、相近实现和验证方式;适用于接手陌生项目或在修改前建立可靠上下文。
-
lovethread Skill Change Implementation在已有代码库中实施范围集中的 Bug 修复或二次开发;适用于目标行为或根因已经足够明确,需要沿用项目模式并保持兼容性的代码修改。
-
easyindie Skill Lark Shared飞书/Lark CLI 共享基础:应用配置初始化、认证登录(auth login)、身份切换(--as user/bot)、权限与 scope 管理、Permission denied 错误处理、安全规则。当用户需要第一次配置(`lark-cli config init`)、使用登录授权(`lark-cli auth login`)、遇到权限不足、切换 user/bot 身份、配置 scope、或首次使用 lark-cli 时触发。
-
dasdigitalemomentum Bundle Review FixApply explicitly accepted, evidence-backed defect fixes in the eligible existing session that produced the review or diagnosis, preserving secured evidence.
-
dasdigitalemomentum Bundle Review PlanIndependent review of a plan against its requirements, scope, and completeness. Produces a structured review with severity-rated findings. Use after create-plan to validate quality before execution.
-
unsafe9 Bundle RescueRescue
-
easyindie Skill Lark Base当需要用 lark-cli 操作飞书多维表格(Base)时调用:搜索 Base、建表、字段管理、记录读写、记录分享链接、视图配置、历史查询,以及角色/表单/仪表盘管理/工作流;也适用于把旧的 +table / +field / +record 写法改成当前命令写法。涉及字段设计、公式字段、查找引用、跨表计算、行级派生指标、数据分析需求时也必须使用本 skill。
-
dasdigitalemomentum Bundle Execute Work PackageExecute a bounded implementation unit from authoritative plan references or an inline gated brief using a stateful subagent loop (steps -> gate -> execute -> digest) without creating new persistent artifacts.
-
dasdigitalemomentum Bundle Review ImplementationIndependent review of a completed implementation against its plan, acceptance criteria, and code quality standards. Produces a structured review with severity-rated findings. Use after execute-work-package to validate implementation quality.
Frequently asked questions
What are Coding & Dev Tools agent skills?
Coding agent skills teach AI agents repeatable engineering workflows: reviewing pull requests, writing tests, refactoring safely, and enforcing house style. Install one SKILL.md and your agent applies the same checklist every time, whether you use Claude Code, Cursor, Codex, or another agent.
Which Coding & Dev Tools skills are most installed?
Popular Coding & Dev Tools skills on SkillMD right now include Questions — will, Real point — load-bearing hinge, Coaching session. Rankings shift as installs change; sort this page by "Most installs" for the live list.
Do Coding & Dev Tools skills work with Claude Code and Cursor?
Yes. Every skill here ships as a SKILL.md file, an open format that works in Claude Code, Claude.ai, Cursor, Codex, Windsurf, and 60+ other agents. Install one with npx skillmds@latest add <owner>/<name>, or copy the file into your agent's skills directory.