team-tasks
📋 Coordinate multi-agent development workflows with a Python CLI tool using shared JSON task files for efficient team collaboration and task management.
コーディングタスク向けの 5,510 件のツールとスキル
📋 Coordinate multi-agent development workflows with a Python CLI tool using shared JSON task files for efficient team collaboration and task management.
Generate a complete, tokenized brand/design system — pick a direction, get tokens, logo, favicon/OG, and guidelines. SDK + CLI + MCP. Productizes the saas-brand-system workflow.
Build, update, debug, explain, review, and document Rust Discord bots or libraries that use discordrs 1.2.0. Use when Codex needs to work on discordrs gateway runtimes, typed EventHandler/Event flows, interactions endpoints, RestClient or DiscordHttpClient helpers, cache managers, collectors, sharding, voice playback/receive, DAVE/MLS hooks, Opus encode/decode, Components V2 builders, modal parsing, examples, docs, or the upstream discordrs project.
Guide for using uv, the Python package and project manager.
generate copilot-instructions.md and micro *.instructions.md files for github copilot code review — repo-aware, stack-specific, applyTo-scoped rules
Parse and read research papers or other document PDFs with MinerU over HTTP APIs using Python requests. Use when Codex needs a reliable paper-reading workflow for local PDFs or PDF URLs, and should choose between fast no-auth Markdown extraction (`flash`) and token-based high-precision extraction (`precision`) with richer layout/assets.
BareScript for Python - a lightweight scripting and expression language
A high-performance DOCX manipulation engine for LLM agents. Features token-efficient reading, surgical editing, template processing, and chart data extraction.
Builds Garmin Connect IQ watch faces in Monkey C — toolchain setup, device capability vetting, always-on (AOD) and sleep handling, performance budgeting, font sizing, app settings, sensor/data sources and permissions, background web fetching, and store publishing. Use when creating, building, configuring, profiling, debugging, or publishing a Garmin watch face or Connect IQ project, or when checking whether a target watch can run a given design.
Use for any React testing task — writing, adding, generating, updating, refactoring, fixing, or reviewing test files (`*.test.*`, `*.spec.*`, `__tests__`) in a React + Jest or Vitest project. Covers unit and integration tests for components, hooks, Redux/RTK, Zustand, React Query, MSW network mocking, forms (react-hook-form, formik), RTL queries, accessibility, and plain JS/TS logic. Auto-detects the app's libraries from `package.json` and generates production-grade tests matching the existing stack.
Reusable AI agent collaboration scaffold system (AGENTS.md / CLAUDE.md / .agents/). Three additive sizes — S/M/L. Works with Claude Code, Codex Desktop, Cursor, Windsurf.
Skill from Restoration/python-clean-architecture
Claude Code skill for AST-aware Python refactoring using rope
Agent Skill for Python code generation, analysis, and debugging using only local context
A Claude Code skill that fixes Chinese quotation marks breaking code syntax. Context-aware typography: ASCII quotes in code, Chinese quotes in prose.
Triangulation-based fact-checking engine for documents. 5W1H extraction, source grading, Python arithmetic verification, LLM weakness compensation.
Use when the agent must read, apply, or review against the official Rust API Guidelines from rust-lang.github.io and must fetch the live pages first instead of relying on memory or local copies. Trigger this skill for Rust API design reviews, crate audits, naming checks, documentation checks, macro guidance, predictability and flexibility reviews, type-safety and dependability reviews, debuggability and future-proofing checks, or any request to consult the Rust API Guidelines online.
ai agent skill for format docx files
Housing skill for AI agents like Openclaw or Claude Code that scrapes Slovenian real-estate websites for rental listings. Provides analysis and ranking based on the users criteria and compiles scheduled reports..
Skill to write code in a static python dialect
Agent skill for Luther substrate/ELPS development — lint, analyze, debug, test via shirotester MCP
当用户要求将Python算法模型打包为docker形式,用于云计算平台或数据汇交时使用此skill。
RStack: Raghav's agentic engineering system with Scrapling and Agent Reach.
Agent-first CLI for generating images from text prompts using Google Gemini
Produce a single self-contained .html file that explains an idea, plan, or piece of code with the fidelity of a designed document — not a wall of markdown. Use whenever the user asks for an implementation plan, PR writeup, system explainer, design exploration, code review, or concept explainer.
A protocol for AI-assisted scientific paper reproduction, numerical model reconstruction, and evidence-driven debugging using structured experiments and Git-backed decision tracking.
Claude Code skill for Fireberry CRM API integration - query syntax, CRUD operations, metadata, field types, and common patterns
Claude agent skill for authoring Power BI PBIP dashboards as source-controlled files — verified template, scaffolder, report-edit CLI, deterministic linter, evals.
コード整理と構造改善のためのリファクタリングスキル。
Claude Code skill for systematically removing unnecessary .clone() calls from Rust code
🎯 Interactive algorithm practice with 100+ problems across Easy/Medium/Hard difficulties. Generates ready-to-run code templates with built-in test cases. Perfect for coding interview preparation and algorithm learning. Covers arrays, strings, linked lists, trees, dynamic programming, graphs, and more.
LLM-compiled knowledge bases for Claude Code + Obsidian. Inspired by Karpathy.
Claude Code skill that audits Figma files against Refactoring UI and Practical UI principles
Skill from souls-syntax/professional
Quiet, editorial chart PNGs for technical blog posts. 15 primitives, six themes, one CLI.
7-lens code review skill for Claude Code -- finds bugs that conventional review misses
Assembly language guardrails, patterns, and best practices for AI-assisted development. Use when working with assembly files (.asm, .s, .S), or when the user mentions Assembly/x86/ARM. Provides calling convention guidelines, register usage patterns, and debugging techniques specific to this project's coding standards.
"通用内容扫描协议 v2.1。确定性规则由 Python 脚本执行,Agent 只负责 Phase 2 语义判断和工具调用编排。支持 Claude Code 和 OpenClaw 双平台运行。"
TypeScript CLI for helpdesk, built with Bun. Agents ready
Run Python programs in the cloud as if they were local! Modal inspired.
An elder brother guiding AI agents to create production-grade apps following strict coding standards, modularity, and database conventions. Heavily optimized for Node.js, React, Next.js, and Python.
Open-source Python skill that audits and optimizes MarkdownFlow teaching prompts for OpenClaw, Claude, and Codex.
Build stateful AI agents and agentic workflows with LangGraph in Python. Covers tool-using agents, branching workflows, memory persistence, human-in-the-loop, multi-agent systems, and production deployment. Supports 14+ AI coding agents via Agent Skill Standard.
State-machine-based TDD workflow skill for Cursor IDE
教 agent 用第三方 Tushare API 代理编写 Python 股票数据代码的 Claude Code skill
기상청 공공 날씨 API용 Python 클라이언트를 구현, 확장, 디버그, 테스트할 때 사용한다. KMA, 기상청, 단기예보, 초단기실황, 초단기예보, 동네예보, VilageFcstInfoService_2.0, APIHub, data.go.kr, nx/ny 격자 변환, base_time 계산, serviceKey/authKey 인코딩, SKY/PTY 코드, apis.data.go.kr/1360000, apihub.kma.go.kr 관련 작업에 적용한다.
Claude skill: StrongAI PID/Aspect compile-time code injection system
Clean Code concepts for Typescript Skill
Drive Windows GUI applications from a shell using two bundled tiny binaries — wmctrl.exe (list windows / switch focus by process name) and xdotool.exe (send keystrokes to the active window, including modifier combos like Ctrl+V, Alt+F4, arrow keys, and literal text). Includes PowerShell helpers for the focus-then-send pattern and a .NET-based active-window screenshot tool that fills in the one capability the upstream binaries lack. Use this whenever the user wants to automate a Windows desktop app from the command line — switching focus between Chrome / VSCode / Cursor / MATLAB / WeChat / any other app, pasting clipboard contents into a foreground window, scripting keyboard shortcuts to trigger menu actions, building "computer use lite" workflows that don't need a full screen-reading agent, or capturing the current foreground window. Trigger on mentions of wmctrl, xdotool, SendKeys, "switch to <app> and …", "focus <app>", "send keys", "paste into <window>", "screenshot the active window", "automate Windows GUI", or any request that maps to focus-window + send-key + maybe-snap.
AI agent skills for Calimero Network development — Rust SDK, JS/Python clients, registry publishing, Desktop SSO, and CLI tooling