mathpix
Turn almost everything into .md for agent workflows, research pipelines, and RAG ingestion.
自動化タスク向けの 7,504 件のツールとスキル
Turn almost everything into .md for agent workflows, research pipelines, and RAG ingestion.
test-creator is a skill that builds a complete, professional test system for any project. It doesn't fix your code — it tells you exactly what's broken and where.
Standalone HCS-26 skill package for Programmable Secrets operator workflows.
|
Guide the user through a deliberate, iterative specification-development workflow for software projects (greenfield, iteration, or adoption of an existing brownfield project). Activates on /spec and its subcommands (/spec setup, /spec interview, /spec adopt, /spec seed, /spec review, /spec revise, /spec check, /spec implement, /spec verify, /spec reconcile, /spec close, /spec decide, /spec defer, /spec help). Maintains artifacts under ./spec/ with an explicit state machine in ./spec/state.yaml.
Use GrayMatter durable memory and graph through api-0 for authenticated reads and writes to MemoryEntry and SwarmOps graph. Use when storing durable decisions, todos, context, artifacts, or querying shared graph/memory state for ValkyrAI and swarm workflows.
Planner/implementer/verifier harness skill for long-running Codex workflows
Scene-aware desktop automation skill for OpenClaw agents on macOS
Claude Code skill: Autonomous PR pipeline from commit to merge
Claude Code plugin: PPTX template ↔ markdown pipeline (parse, generate, revise)
Claude Code skill for authorized offensive security workflows: recon, stealth automation, hidden-API discovery, incremental sync. 4 playbooks + 29-tool reference.
Turn an AI coding agent into a reliable Zama FHEVM developer for confidential smart contracts and dApps. Use when an agent must design, implement, audit, test, self-correct, and deploy FHEVM Solidity, Relayer SDK frontends, input proofs, encrypted types, ACL/decryption flows, ERC-7984-style confidential finance, x402-style encrypted access fees, RWA lending, remittance, payroll, or compliance-native private finance workflows.
OpenClaw skill for remote Playwright browser automation
Redux-shaped orchestration for long-running AI coding loops
An OpenClaw Skill that orchestrates Claude Code as a background engineer — plan, approve, execute, and validate coding tasks without blocking your Agent.
"Run a comprehensive, multi-phase evaluation of any idea, product, or feature set — producing multiple deliverable files covering critical feedback, competitive context, strengths & opportunities, design issues (including optional live-site visual/UX audit when an HTTPS product URL is in scope), optional security posture audit (Phase 1E), optional AI/agent durability audit (Phase 1F), proposed fixes, and a content strategy outline. Accepts any combination of raw text, uploaded documents, or crawlable URLs as input. Orchestrates parallel subagents for efficiency. Use this skill whenever someone says 'evaluate this idea', 'tear down this product', 'full analysis of', 'mega eval', 'comprehensive review', 'deep evaluation', 'assess this feature set', 'product review pipeline', or any request for a thorough multi-dimensional analysis of a concept, product, or feature. Also trigger when the user provides a URL, doc, or text block and asks for a complete picture — strengths, weaknesses, market context, and next steps."
helps ocean science researchers turn proposals, figures, code, data descriptions, target-journal plans, Zotero-supported literature, and advisor feedback into staged manuscript materials through prepare, methods, structure, writing, review, and polish workflows.
Reddit automation skill collection for AI agents — post, search, comment, interact via browser extension bridge
Manage a family bookkeeping workflow backed by Feishu Bitable. Use when the user wants to record an expense or income, say things like "记一笔""记账""入账""查本月支出""看年度统计""修改这笔账""删除这笔账", or import/export household transactions from WeChat or Alipay bills. Also use when the user wants monthly or yearly summaries, category breakdowns, member-based spending analysis, duplicate-safe bill imports, or natural-language CRUD for a shared family ledger.
MCP server for OpenClaw - expose smart home, memory, and automation tools to Claude Desktop, Cursor, Windsurf, and any MCP client
個人財務追踪系統 (含 n8n) Demo
Claude Code + Codex dual-agent orchestration skill with caveman token compression and darwin self-improvement
Create, test, and optimize multi-agent workflows for OpenCode. Use this skill whenever the user wants to build a team of AI agents, design an agent pipeline, create an orchestrator with specialists, set up coordinated agents, or optimize an existing OpenCode workflow. Also triggers when the user says things like "I want agents that work together", "build me a workflow", "create a DEV agent", or "optimize my agents".
An AI translation pipeline for high quality light novel translations
Spin up and manage a collaborative coding team using Claude Relay. Foreman turns multiple Claude Code sessions into a coordinated crew with defined roles (Orchestrator, Dissenter, Workers, Cleaner, Circuit Breaker) that communicate via natural language to build software right the first time. Triggers include "spin up foreman," "launch the crew," "start a coding team," "foreman build," "staff the job site," or any request to coordinate multiple Claude Code agents for a coding task. Also use when the user says "ask the dissenter," "broadcast to the team," "check on the workers," or references Foreman roles by name. Use this skill whenever multi-agent coding collaboration is needed, even if the user doesn't say "foreman" explicitly.
AI可执行剧本检查表V3：7-stage AI剧本检查、诊断、改稿与标准化 Agent Skill
Assistant expert en développement logiciel et DevOps conçu pour générer du contenu technique rigoureux (messages de commit, issues, PR, release notes) et standardiser la gestion de projet Git selon les meilleures pratiques de l'industrie.
扫描 _now.md 未处理节 → 跨参照项目文档 → 分类搬运策划方案 HTML。触发：/capture-loop 或 codex exec --yolo $capture-loop (cron 每日 21:00)
"Use when you need to retrieve historically accumulated knowledge, methods, or behavioral patterns before answering or acting; when a conversation or workflow produces durable insights worth preserving for future reuse; or when you discover existing wiki content that is outdated or incorrect and needs correction."
Claude Code skill for building pixel-perfect websites in Webflow Designer using MCP Bridge and browser automation
Real-time cryptocurrency prediction system powered by multi-source data fusion, technical analysis confluence scoring, and deep learning.
Agent-first job runner for API workflows.
Automated LinkedIn prospecting & job search platform. Browser automation + Claude AI scoring across LinkedIn, Naukri, and Hirist. Dashboard with filtering, exports, and outreach drafts.
Dynamic multi-agent orchestrator skill for Claude Code. 22 specialized agents, 7 phases, 5 pipeline types.
AI research advisor — 11-phase pipeline from literature survey to paper submission. Acceptance scorecard, claim-evidence tracking, simulated peer review. Claude Code skill.
Heady Monorepo — Continuous Semantic Logic · Latent Intelligence OS · 17-Swarm Orchestrator
OpenBSD ports workflow grounded in OpenBSD docs and man pages. Use when working in /usr/ports or a local ports tree, building packages, or submitting new ports.
AIエージェントを並列で動かすためのSKILLS
TPVL (Taiwan Professional Volleyball League) skill for OpenClaw - scores, schedules, standings
三层多角色协同任务编排系统 - Layer 0/1/2 Multi-Agent Task Orchestration System
Deep research and browser automation via remote Chrome instances using CDP. Use this skill for in-depth research across multiple platforms (Google, X/Twitter, Reddit, Quora, Medium, academic sources), web scraping, browser automation, form filling, and any task requiring real browser interaction. Trigger when the user wants to research a topic, find recent information, scrape data, automate web tasks, or interact with websites. Even casual requests like "research X", "find recent posts about Y", "what are people saying about Z" should trigger this skill.
Human-in-the-loop LinkedIn engagement workflow for the native Windows or macOS Codex app using the Codex Chrome plugin/extension for the user's signed-in Chrome session, with @computer only as a visible native UI fallback. Use when asked to review LinkedIn feeds, search results, profile activity, company pages, or post URLs; identify relevant posts; avoid duplicate reactions or comments; draft contextual comments; and perform only explicitly approved visible UI actions through the logged-in browser session.
Master coordinator for complete T180 (180nm) IO Ring generation. Handles signal classification, device mapping, semantic intent generation, engine-based pin wiring, JSON generation, SKILL generation, and complete workflow through DRC/LVS verification. Use this skill for any T180 IO Ring generation task. Trigger when user mentions T180, 180nm, 180nm IO ring, or any IO ring task targeting the 180nm process node.
WeChat Article Writer: end-to-end 微信公众号 publishing pipeline
Automated prospect-to-website pipeline for trades businesses. Discover, score, build, and sell websites using AI.
AI Orchestrator skill for Claude Code — routes coding and analysis tasks across AI CLI tools
Task generation and breakdown workflow. Use when decomposing a PRD or complex feature into implementable subtasks with clear acceptance criteria and dependencies.
Multi-model skill testing and vetting workflow for OpenClaw
Codex skill for GRL-style mechanism-centered scientific writing workflows
Concept-first writing pipeline that drafts, evaluates, and revises in fixed order (concept → structure → line). Bring your own voices.