Event investigation, decision reuse and authorized action workflows.
Directory
Browse agent skills
Search by intent: design systems, testing, deployment, security, documentation, Azure, Supabase, React, or any workflow you want an agent to perform better.
Showing 3,751–3,780 of 125,958 skills
Page 126 of 4,199 · loading 30 at a time
Check how much of a website AI assistants can actually read: JavaScript-only pages, robots.txt rules for AI crawlers, Cloudflare blocks, structured data, llms.txt and more.
Build, check, simulate and preview store games on table-games-engine.
The support window of Mapled, the hosted headless CMS for sites built with AI — a chat button on any site: visitors write to the project's team, answers come back in the window and by email. One script tag, or a React component.
Builds a professional website from a conversation and publishes it to the person's own domain and hosting, for people with no technical knowledge. Use when someone wants a website made and put live on their own domain, wants an existing site replaced, or someone is building a site on another person's behalf, also helps with publishing/hosting
Local gateway that routes coding-agent tool decisions through Jev and can attach a read-only advisor (GPT or Claude).
Product-management workflows, specs, and verification for coding agents.
Free Cursor rules + AGENTS.md that stop AI (Cursor, Claude Code, ChatGPT) from writing broken Unity C#: missing usings, invented APIs, Unity 6 vs 2022 mixups. MIT.
Official Vericto SDK for Node.js: check every SQL statement with Vericto before it runs. Wrappers for snowflake-sdk and better-sqlite3, LangChain.js SQL agents, and a client for the Vericto Runtime API.
Document designer for Kumiko: template overview, layout editor (mail layout, PDF letterhead) and the PDF template editor.
Claude skill that audits web apps, APIs, and mobile apps (Android/iOS/hybrid), fixes what is safe to fix, and explains the risk in plain English. OWASP Top 10, API Top 10, MASVS/MASTG, and ASVS.
Roku typescript testing library
Deterministic Playwright flows that double as narrated, branded videos, with a development-then-staging release gate.
Dynamic subagent model selection, per-call context limits, and gateway auto-sync patcher for Claude Code (9Router, LiteLLM, OpenRouter)
Animated, copy-paste UI components for Expo and React Native — free components, CLI and expo-charts
Bonsai: a git platform for many agents working at once, on Workers and Artifacts
ipx - the InferencePort AI command line (cloud, organizations, local models, host server). Ships a native binary; no Python needed.
Model auto-ranking and transparent failover for DeepSeek Harness: ranks discovered models by measured connectivity and speed, switches to the next candidate inside the same turn when the current one is unusable, and ships a grouped model list with per-cha
输入框提示词联想插件:助手回复结束后预测你的下一句话,以灰色 ghost text 浮在输入框里,Tab 采纳、Esc 忽略 | Codex-style composer prompt predictions for DeepSeek Harness: your next message ghosted into the composer, accepted with Tab.
Temporary package placeholder for staged publishing
TypeSafe-compatible question and answer client backed by AI SDK structured output
Claude Code skill: keep Claude's questions and sample images on one artifact page and answer by pasting (決めごとページ)
TeamSpeak SDK for TypeScript: an independent voice client for Node.js, and the typed commands, ids and errors every connection shares
filegoblin PDF vision model: text recognition for scanned pages (PP-OCRv6 medium, Apache-2.0) (install @filegoblin/models instead)
Notifies you when DeepSeek Harness needs attention: approval prompts, questions, task completion, and errors — with a Windows toast whose button brings the window back.
Liumi VectorCore (.lvc) - sub-2ms SIMD INT8 vector search, zero-copy on-device AI
Official TypeScript Core SDK for Sarvam AI Indic language models and speech services
Indra OCR running locally in browser WebAssembly workers