Engineering plugins

A growing collection of agent skills covering software engineering, mobile release engineering, and app store publishing.

3 skills

Divan's agent skills, packaged as a Claude Code plugin.

3 skills

Java 测试套件（2 技能）：不教测试代码怎么写，教什么时候绿灯是假的。开发期纯单测（AssertJ 取值、mock 边界、三条假绿检查点）+ 集成测试（回归谁值守 / 放哪层 / 环境用什么跑 三决策，路径不对 / 时间点不对 / 失败不算失败 三机制）。

2 skills

MyBatis-Plus（baomidou）ORM 增强框架开发助手——CRUD、分页、条件构造、Mapper/Service 层、逻辑删除与乐观锁。项目已引入 MyBatis-Plus 时使用。

1 skill · hooks

Sa-Token（cn.dev33）权限认证框架开发助手——登录鉴权、权限角色、SSO、JWT、会话与接口保护。项目已引入 Sa-Token 时使用。

1 skill

Read-first GitHub Rulesets governance pack for ForgeRail.

1 skill

Project-runbook-first release governance pack for ForgeRail.

1 skill

Design a PCB end to end: atopile circuit, KiCad layout, Freerouting autorouting, DRC, gerbers, BOM and pick-and-place for fabs such as JLCPCB.

1 skill

Matthew Boston's personal software engineering skills for coding agents.

60 skills · 1 agent

Skills for React/React Native Engineers — clean code, strict TypeScript, and native-feeling animations (GSAP, Reanimated, Skia). Straight from my .claude directory.

38 skills · 8 commands · 8 agents

Passionate Quantum Absence: superpose divergent branches, attack them, collapse on evidence. Fable 5 where quality is decided.

12 skills · 12 commands · 14 agents · hooks

Core Ravn agent plugin. Bundles reusable skills and MCP server configuration.

33 skills · 2 agents

Repo-local Codex plugin exposing focused AHKFlowApp project skills.

27 skills

Connect OpenAI Codex CLI to McpServer for session logging, TODO management, requirements tracking, GraphRAG operations, Android device validation via adb_step, and per-turn enforcement scripts.

18 skills

Nine performance engineering skills: CPU optimization, memory leak detection, race condition testing, database query analysis, chaos engineering, network payload reduction, CI/CD acceleration, telemetry, and UI frame-rate enforcement. Every skill requires benchmarked proof before presenting changes.

17 skills

Thirteen skills (start, begin, scope, spec, stack, architect, design, verify, code-review, debug, deliver, maintain, comply) that lean on Claude Code's bundled /verify, /code-review and /security-review for the running and the bug hunt, three review agents on Sonnet, and the guard that refuses a bypassed gate or a force-push before it runs.

13 skills · 3 agents · hooks

Cross-host development workflow: setup, dependency updates, checks, review, prose hygiene and documentation sync, PR follow-through, and branch promotion with conditional release verification.

8 skills · 6 agents · hooks

Orchestrate issues across models, merge PRs one deploy at a time, hand off between threads, verify reviews, audit apps and sites, write European Portuguese, and guard Laravel databases.

13 skills · hooks

project-map mapeia o repositório para dar contexto real às demais: referências de arquitetura (NestJS, monorepos), boas práticas React/Next.js e o fluxo sdd + spec-harness (specs testáveis com TDD e enforcement de path) ou spec-orchestrator (implementação lean via subagentes).

12 skills · hooks

Browser automation CLI for AI agents - interact with and test code in Chromium using natural language

10 skills

Comprehensive PR review with specialized agents for code review, silent failure hunting, and test analysis

1 command · 6 agents

通用代理工作流：釐清需求、規劃、驗收與交接；個人政策由外部模組明確啟用。

5 skills · hooks

Complementos ao fluxo do mattpocock-skills: portão de estado compartilhado antes do fatiamento, checkpoint de consistência a cada 5 tickets e orquestrador autônomo da fila.

3 skills · 1 agent · hooks

Canonical React UI/UX aesthetic + atomic-structure rules with a planning architect agent, an implementation agent, and plan/review-ui skills

3 skills · 2 agents

Learn from the current session or from recurring history across sessions, git changes and review feedback, and propose evidence-backed skill and guidance improvements.

2 skills · 2 commands

Whose Turn, the Claude Code side of a local monitor of your Jira, GitHub, Linear and other sources: whose turn it is, what changed, your own to-dos, a summary at the start of each session, a hand-off from the monitor window into the session, a /whose-turn command that opens the window, and an MCP server for any agent. The monitor itself is the published whose-turn npm package; the plugin installs its latest version into the monitor's data directory on first start and runs it from there; each newer version is then announced in its window and at session start and installs only when the user says yes or clicks Update. Employer-specific values live in a workspace config on the machine, not in the plugin.

1 skill · 1 command · 1 MCP server · hooks

CERBERUS — the two-headed guard at the gate for a PUBLIC skills marketplace. A PostToolUse hook fires on every skill / reference / script / eval / manifest edit and reminds you to run BOTH heads before committing: cerberus:leak-check (outward — no real work identifier or secret escapes into the public marketplace) and cerberus:security-scan (inward — the eight-point consumer-safety checklist: prompt injection, exfiltration, secrets, dangerous commands, obfuscation, external fetches, credential access, privilege escalation, so nothing unsafe reaches whoever installs the skill). Before a push it also guards the commits themselves — their messages, authors and branch name, which no file check reads: a classic PreToolUse hook has a nested claude judge the outgoing commits and blocks a push that leaks or that it cannot judge, and on Claude Code 2.1.292 or newer a push-guard mod judges them in-process first; an older Claude Code gets a red notice at session start to update. The verdict cache is guarded best-effort, so the hook stops a forgotten check, not a deliberate forgery. Both are agent judgment passes, no denylist by design — a list of the real names to catch would itself be the leak; the intelligence lives in the agent, not a regex.

2 skills · hooks

Passive skill that reads the compiled catalog to discover/recommend/install marketplace skills, surfaces declared side effects, and treats catalog text as untrusted data.

2 skills · hooks

Manage Claude Code hook rules; Codex has no hook adapter.

3 skills

Consult Claude Code as a second-opinion agent for brainstorming, plan validation, and code review. Aliases: @claude, @claude-code.

1 skill · 1 MCP server