Claude Code and Codex plugins their authors filed under "development". All categories.
Production-grade engineering skills for AI coding agents — covering the full software development lifecycle from spec to ship. (Original de Addy Osmani, 98.7K★, MIT, fijado al commit auditado 1401c8b — PASS con condiciones, 2026-09-23 y re-auditado 2026-10-10: preguntar antes de reset --hard, push, deploy, npm/brew/pip install o agregar MCP.)
25 skills · 4 agents
The always-on engineering rules, core skills, and the tier and domain specialist agents. Requires the repo-docs plugin for repo grounding.
11 skills · 13 agents · 4 MCP servers · hooks
스펙부터 배포까지 개발 생애주기 스킬 (외부: addyosmani/agent-skills, MIT)
25 skills · 4 agents
Start with /getting-started — an interview-first concierge that sets newcomers up: it checks your config, adapts to your subscriptions, and recommends the right tools. Also includes: claude-md-optimizer, unify-agents-md, project-documenter, technical-documenter, feature-documenter, raginclude-generator, model-registry, astrocade-game-creation. For version-manager, repo-doctor, update-code-map, and yt-transcript (which ship bundled tools), install repo-hygiene / docs-toolkit / yt-transcript from the Chameleon Labs marketplace (Chameleon-Labs-LLC/plugins) — or clone this repo and follow the README.
29 skills
Engineering workflows distilled from real Claude Code sessions — ship-pr, pr-triage, stack-surgery, gh-stack, deep-pr-review, pr-review-graph, code-style, worktree-create, worktree-dev-server, dev-doctor, verify-in-app, bug-intake, warehouse-data, prod-script, vendor-api-probe, render-pipeline-debug, firefighter-loop, cloud-agent-work, cloud-runtime, demo-video, sandbox-tool, research-doc, visual-explainer, one-pager, slack-message, status-update, daily-summary, agent-team, mine-sessions
29 skills
AI-native project architecture — structured context, enforcement hooks, multi-agent orchestration, and 32 grouped MCP tools (more on demand) for symbol-driven development
21 skills · 5 agents · 1 MCP server · hooks
Evennia MUD game development skills covering typeclasses, commands, attributes, rooms, exits, scripts, web APIs, and admin workflows.
28 skills
Solo developer SDLC framework for Claude Code — spec-first, TDD, automated testing, brain-integrated
11 skills · 14 commands · 2 agents · hooks
Enterprise AI agent orchestration plugin with 150+ commands, 74+ specialized agents, SPARC methodology, swarm coordination, GitHub integration, and neural training capabilities
25 skills · 3 MCP servers
Enterprise AI agent orchestration plugin with 150+ commands, 74+ specialized agents, SPARC methodology, swarm coordination, GitHub integration, and neural training capabilities
25 skills · 3 MCP servers
Focused, progressively disclosed language and engineering-method skills for repository-adaptive software development, including concurrency and runtime-specific guidance.
27 skills · hooks
Author + operate Stargraph — graphs, skills, nodes, tools, triggers, Bosun rule packs, markdown SKILL.md skills, directory-based plugins; run/replay/counterfactual against stargraph serve, drive HITL, inspect checkpoints.
2 skills · 18 commands · 8 agents
Ultimate AI software team — 17-stage spec-anchored pipeline with dual interrogation (PM + design), parallel research, locked tests, adversarial Ralph Loop, anti-cheat gate, inter-stage review, spec GraphRAG, Reflexion lessons, failure recipes, cheap-model triage, and Phase 3 CI/CD ensemble reviewers.
2 skills · 8 commands · 17 agents · hooks
Universal source control operations across GitHub, GitLab, and Bitbucket with modular handler architecture
28 skills
Operational log management with session capture, hybrid retention, archival, search, and analysis
28 skills
Frontend design skills. Build: emil-design-eng (start here for components and screens), landing-page-design (single-CTA conversion landing page with a visual system), build-awwwards-quality-sites (immersive scroll-storytelling marketing sites), apple-design (Apple HIG principles). Motion: animate (add or refine animation on an already built site), animation-vocabulary (turn vague motion words into exact animation terms), motion (Framer Motion implementation + discipline), motion-system. Review and fix: screen-critique with the /critique-screen command (judge a rendered screen against seven critique lenses), interface-review (review UI changes in a PR, branch, or working tree with regression detection), better-interface (the holistic consolidation and verdict layer over the better-* family; uninstalled domains report as Not reviewed), better-layout (spacing, alignment, and grouping to the pixel), perception-laws (Gestalt laws, Von Restorff, visual hierarchy), accessibility-audit (WCAG 2.2 POUR). Systems: design tokens, typography scale, shadcn/ui composition, stitch-skill (create/update DESIGN.MD), taste-skill (targeted visual polish pass). Bridges (not vendored): impeccable-bridge for the full end-to-end UI/UX pass via /pro-dev install, taste-skills-bridge for the Leonxlnx anti-slop, visual-style, image-gen, and brand-kit library via /pro-dev install, plus web-design-engineer-bridge (reference-grounded HTML artifacts, style recipes, design scoring), tastemaker-bridge (extract tokens from a reference image's real pixels and lock the style across a project), and mengto-skills-bridge (scroll storytelling, WebGL and shader effects, video-to-superprompt) - all three installed with /pro-dev install. Install figma from claude-plugins-official via install-companions.sh. Bridges: react-patterns-bridge (Vercel composition and view-transition patterns). Lint: web-design-guidelines (file:line check of UI code against Vercel Web Interface Guidelines at a pinned commit). /pro-des
26 skills · 2 commands
Python development with Django, Flask, FastAPI, async patterns and uv for package management
24 skills · 1 command · 3 agents
Turns what a client says into documents they can correct, a machine spec an agent can build from, and a work list somebody can pick up. Keeps the macstack/ folder of a project: macstack.json — the standardized business + technical stack specification, always English — and the six client documents it is written from. OVERVIEW says what the product is and who it is for; USER-CASES carries each case with its UX bar and an addressable acceptance list; UX-UI states what each screen shows and what must never appear on it; AUTOMATION is the trigger -> task -> workflow -> role model; HANDBOOK is how a person actually uses the thing; OPEN-QUESTIONS splits what the client owes from what the team deferred. v3 makes those six pure markdown — headings and bullet lists, nothing else. No YAML blocks, no tables, no change-log sections: the only machine markup is an HTML comment the reader never sees, pointing each entity at its place in the spec. A client can edit the document in any editor and hand it back. Around them: an immutable inbox for anything a client sends, a gated delta/rulings loop that merges it, generated requirements, architecture, test cases and index that carry every id the client documents carry, an append-only ledger with one row per edit and per client comment, tasks reconciled with the team's own tracker, and a review package that shows each statement with its own history and reads the client's answers back into the ledger. Eight commands, one job each — including one that reconciles the whole folder against the source tree in a direction you have to declare: the code is master and every document is corrected through a gate that never silently overrules an answer the client gave, or the documents are master and the gaps become tasks. Every edit is journalled, every finished task sweeps the client documents and not only the generated ones, task statuses move to what the audit actually found — closing what is built and reopening what is not — and every document
18 skills · 8 commands · 1 agent · hooks
Equipo de Marketing, Ventas y Publicidad con 8 agentes IA para vender un proyecto, una propuesta o una mejora: nunca una presentacion generica, sino el producto DESGLOSADO PARTE POR PARTE. Primero el mapa del producto con una ficha por parte (que es, problema y su costo, como funciona con pantallas reales, que gana el cliente con fuente, antes/despues, benchmark) aprobado por el usuario; luego las piezas: video keynote cinematografico con un capitulo por parte y clips por parte (HyperFrames), deck de propuesta comercial con estetica de lanzamiento, one-pager, panfletos y redes sobre la marca FAAST. Investiga mercado, competencia y tendencias con fuente y fecha. Nada inventado, producto real, aprobacion por pieza. Punto de entrada: /marketing:start.
7 skills · 11 commands · 8 agents · 1 MCP server · hooks
Deslops AI-written code. Always-on standards and per-phase skills for coding agents.
22 skills · 5 agents · hooks
Puspesh's personal Claude toolkit: skills, subagents, and commands.
16 skills · 1 command · 10 agents
Canonical Vibe Engineering House Skill plus supporting procedures. The umbrella skill owns ICM context, INTENT → SPEC → PLAN → BUILD → PROOF → COUNCIL → JUDGE → SHIP → OPERATE and routes to smaller skills only when needed.
20 skills · 7 commands
仕様策定・実装計画・Issue作成支援プラグイン
12 skills · 14 agents · hooks
Comprehensive codebase analysis and steering context generation for AI agents. Automatically detects project type and generates architecture documentation, design patterns, quality reports, and AI-ready context files.
7 commands · 20 agents
Build, train, evaluate, and use open source AI models, datasets, and spaces.
26 skills · 1 MCP server
Build, train, evaluate, and use open source AI models, datasets, and spaces.
26 skills · 1 MCP server
Build, train, evaluate, and use open source AI models, datasets, and spaces.
26 skills · 1 MCP server
Build, train, evaluate, and use open source AI models, datasets, and spaces.
26 skills · 1 MCP server
Build, train, evaluate, and use open source AI models, datasets, and spaces.
26 skills · 1 MCP server
Turns what a client says into documents they can correct, a machine spec an agent can build from, and a work list somebody can pick up. Keeps the macstack/ folder of a project: macstack.json — the standardized business + technical stack specification, always English — and the six client documents it is written from. OVERVIEW says what the product is and who it is for; USER-CASES carries each case with its UX bar and an addressable acceptance list; UX-UI states what each screen shows and what must never appear on it; AUTOMATION is the trigger -> task -> workflow -> role model; HANDBOOK is how a person actually uses the thing; OPEN-QUESTIONS splits what the client owes from what the team deferred. v3 makes those six pure markdown — headings and bullet lists, nothing else. No YAML blocks, no tables, no change-log sections: the only machine markup is an HTML comment the reader never sees, pointing each entity at its place in the spec. A client can edit the document in any editor and hand it back. Around them: an immutable inbox for anything a client sends, a gated delta/rulings loop that merges it, generated requirements, architecture, test cases and index that carry every id the client documents carry, an append-only ledger with one row per edit and per client comment, tasks reconciled with the team's own tracker, and a review package that shows each statement with its own history and reads the client's answers back into the ledger. Eight commands, one job each — including one that reconciles the whole folder against the source tree in a direction you have to declare: the code is master and every document is corrected through a gate that never silently overrules an answer the client gave, or the documents are master and the gaps become tasks. Every edit is journalled, every finished task sweeps the client documents and not only the generated ones, task statuses move to what the audit actually found — closing what is built and reopening what is not — and every document
26 skills · hooks