rapidly
claudecodexTest the idea before you build it. Rapidly turns an idea into a Lean Canvas, a measurable hypothesis and Pretotyping experiments, then returns a build prompt for the smallest test worth running.
4 skills · 1 MCP server
Claude Code and Codex plugins their authors filed under "productivity". All categories.
Test the idea before you build it. Rapidly turns an idea into a Lean Canvas, a measurable hypothesis and Pretotyping experiments, then returns a build prompt for the smallest test worth running.
4 skills · 1 MCP server
Add AnnotateKit to an iOS app and turn precise on-screen annotations into verified Codex code changes.
2 skills · 2 commands · 1 MCP server
宿主原生多 Agent 协作与独立网页追踪
5 skills
只读地读取各 AI 客户端留在本机的会话与原生记忆,五个客户端各一个技能。
5 skills
Cadence — plan dependency-aware parallel development cycles (cadence-planner) and drive them autonomously to merged PRs (cadence-executor). Requires the superpowers plugin; optionally accelerates analysis with the graphifyy CLI.
2 skills · 3 commands
Talk with your agent: spoken status lines, listening, dictation and conversation, on your machine.
5 commands
Game-development skills on the -smith motif. Five members: pixelsmith (pixel art that reads at several zoom scales), godotsmith (Godot 4.x conventions and proof a build is green), soundsmith (game audio judged by numbers), slicesmith (the test-first coding loop), uismith (menus, HUD and dashboards with a fit probe at every width and theme). Standard-library scripts allowed when declared; Playwright is uismith's one optional package.
5 skills
Content and business pack: video scripts, articles, content plans and unit economics.
3 skills · 2 agents
explain-with-slides: turns a codebase or a pull request into a walkthrough deck with screenshots from the running code, UML diagrams, key files, code and diff slides, tech stack and lines of code. explain-with-video: turns the same walkthrough into a narrated MP4 with captions, chapters and a transcript. refactor-tests: brings a file, class or branch's tests up to written rules for each test type, and covers every case. self-documented-code: renames everything in a scope so each name explains itself and each concept has one name. dont-repeat-yourself: finds concepts implemented more than once and gives each one well-tested implementation.
5 skills
Review pages for agent work: write a parley document, serve it, and read the reviewer's answers back.
5 skills
Edit talking-head videos into Reels, make carousels and images as JPEG, and post anything through Dandelion. Runs on your computer.
3 skills · 1 MCP server · hooks
Write, publish, research and engage on note.com from AI coding agents via a bundled unofficial-API CLI
5 skills
Generate images and videos and work with Toonkit canvases through MCP.
3 skills · 1 MCP server
Synchronize Salesforce logs, find the right transaction locally, and investigate failures and performance with evidence.
3 skills · 1 MCP server
Mem0 memory layer for AI applications. Add persistent memory, personalization, and semantic search to Claude workflows.
2 skills · 1 MCP server · hooks
Skills for working with Nodel distributed control framework
4 skills
Learns your working style and negotiates an update with you when what it actually does drifts from what you've declared.
1 skill · 1 agent · 1 MCP server · hooks
Run setup, research, and implement modes against PLAN.md, TODO.md, and TESTING.md.
4 skills
Live two-way Matrix ↔ TUI bridge via Channels API. Always-on daemon + per-session routing + headless `claude --print` fallback.
2 commands · 1 MCP server · hooks
Add local speech input and output without adding tool privileges.
3 skills · 1 MCP server
Agent skills for the SupSub CLI: auth / sub / search / mp
4 skills
Back up and restore Claude Code configuration via the claude-backup CLI.
4 commands
Generate standup summaries from GitHub activity - personal, team member, or whole team
1 skill · 3 commands
Persistent, provenance-aware memory and passive trace capture for Claude Code.
2 skills · 1 MCP server · hooks
Mine past sessions for corrections; distil them into per-repo and global rules with verifiable evidence.
1 skill · 2 commands · hooks
The entire Slack Web API for Claude Code — 201 methods, approval-gated sending, a cached workspace reference, and scheduled-job alerting.
2 skills · 1 command · hooks
Long-term memory: remember / ask / status, with passive capture hooks (SessionEnd / PreCompact / context-pressure).
2 skills · 1 MCP server · hooks
Markdown 기반 auto memory 시스템 — 프로젝트 간 지식 축적을 위한 4가지 메모리 타입(user/feedback/project/reference) 자동화 커맨드 모음
4 commands
A visually enhanced Claude Code statusline with Nerd Font icons and 256-colour gradient bars, powered by OMC data modules.
1 skill · 2 commands · hooks
Terse markdown research-plan-implement workflow
4 skills