Progressive-disclosure instructions an agent loads on demand. Cheap until invoked, which is why the per-session figure on each card is usually two digits.
A command for reviewing the current state of a SpecPilot requirement, where SpecPilot tracks feature requests through planning and implementation stages. It checks the request against its stage and reports what is complete or still missing.
Spawn one or more OpenAI Codex CLI subagents from bash to offload context-heavy work from the parent Claude. Use when the user says "use codex", "build with codex", "ask codex", "get a codex second opinion", "compare codex vs claude", "have two agents try it", or otherwise asks to delegate or parallelize work to…
Generate an agent skill from a GitHub open-source repository. Use when asked to create a skill from a repo URL, turn a GitHub project into a usable skill, or generate skill files for a CLI/library/framework. Triggers on phrases like "create a skill from this repo", "make a skill for vllm", "turn this GitHub project…
Apply or audit DTS (Dense Technical Syntax), a standard for token-dense English. Use to write or rewrite any persisted text — READMEs, docs, code comments, commit and PR bodies, checklists, error messages, CLI help, tool descriptions, and agent prompts. Also use when asked to de-slop, tighten, compress, cut verbosity…
Dispatch a research or build task into a live tmux session running another LLM CLI (Codex/GPT, GLM/Z.AI, Claude, etc.), then collect the result via a file-based output contract and an event-driven handback ping. Use when the user wants to hand work to another model running in a tmux window, fan a long-running task out…
A Chinese-language image-creation skill for turning an article's key idea, process, or metaphor into hand-drawn illustrations. It uses a consistent brown-haired, round-glasses character on a mostly white 16:9 canvas.
A development procedure for adding features to a TaskBoard application using TDD, or test-driven development. TDD means writing a failing test first, then implementing the feature and checking it in the browser.
Build a user-controlled, evidence-traceable job-search pipeline for unemployed or underemployed people: cross-check job-market signals, map truthful experience and portfolio evidence to roles, tailor resumes and cover letters, maintain an application board, draft follow-ups, and review feedback. Use when the user is…
Use this skill when configuring or operating the mcp-server-mysql MCP server for MySQL access, including generating Claude Code, Cursor, or VS Code MCP configuration, choosing safe permission settings, validating connectivity, and explaining available MySQL MCP tools.
Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.
Use RoutIR — call a running endpoint with the client, stand up a local server that mixes locally-hosted services with services proxied from a remote master server, and extend RoutIR with new bi-encoders, rerankers, and document collections by writing a small config (and optionally one Python file referenced via…
Creates structured multi-file implementation plans before build. Scope-aware: trivial tasks exit early, simple tasks get fast plans, complex tasks get full research + phase files + validation. Enforces Gate 1. Activated by /mk:plan or /mk:cook. NOT for ticket complexity analysis against an existing codebase (see…
Entity intelligence and founder clarity inside Claude Code. Search any company, run banker-grade deep diligence, get actionable remediation steps, and inject company truth into your session. Delegates implementation tasks to Codex when available.
Platformer Movement in FlatRedBall2. Use when implementing platformer mechanics including jumping, ground detection, PlatformerBehavior, PlatformerValues, double jump, air control, variable-height jumps, or side-scrolling movement. Trigger on any platformer-related question.
A read-only monitoring workflow for applications deployed on mainland Alibaba Cloud. It reports availability, performance, server and database status, and estimated costs in Chinese yuan.
A Human Design reading workflow that uses a person's birth date, exact time, and birthplace to calculate and explain a BodyGraph. A BodyGraph is the chart used by Human Design to describe its system of types, centres, channels, and other traits.
Design and verify tools that AI agents can actually use — for any framework or language (MCP servers, LangChain/LangGraph, function-calling, raw JSON schema; TypeScript, Python, or otherwise). Use when writing a new tool for an agent, reviewing or fixing an existing tool definition, deciding how to split capabilities…
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: