Manifest-driven IaC for Salesforce Data Cloud / Agentforce — declare connections, data streams, DMOs, mappings, and search indexes in TypeScript and deploy idempotently.
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 37,771–37,800 of 129,642 skills
Page 1,260 of 4,322 · loading 30 at a time
General-purpose agent with transport abstraction, state management, and attachment support
Claude Code skill: dispatch three calibrated reader personas in parallel against any draft and consolidate feedback into a prioritised editing report
Check Point AI Security Official JS/TS SDK (Workforce AI + Browse Security)
Patchright wrapper with optional Turnstile auto-click helper.
Claim XiaShe external Agent feeds and safely receive protected Skill packages.
Global AgentPie CLI. It shares the core SDK, Agent, Skill, connectors and runtime implementation with `xiashe`, but defaults to:
`xiashe` exposes XiaShe core workflows from a local terminal or any shell-capable Agent.
Build something meaningful, on Sui. A superpower for AI coding agents (Claude Code, Codex, Cursor) to ship real products on Sui.
Comprehensive SEO + AEO + GEO audit skill for Claude Code. Real-browser Core Web Vitals, Schema.org validation, live LLM citation tracking across ChatGPT/Perplexity/Claude/Grok, internal link graph, hreflang, multi-page audits, growth opportunities. Works with any website.
Crawl a website and emit a valid sitemap.xml — for sites that ship without one or whose CMS pipeline forgot to.
Curated, sourced list of AI crawler / training bot user agents, plus a small CLI to test whether a URL is reachable to each bot.
Give AI coding agents a voice. Cross-agent skill that wraps OpenAI gpt-realtime-2 for Claude Code, Hermes, and OpenClaw.
Telegram bridge daemon for Agent Relay
JDI (Just Do It) — lean workflow toolkit for Claude Code, GitHub Copilot, Antigravity, and OpenCode. 10 commands (7 loop + ralph + adopt + meta), 6 core agents + N per-project specialists with file-glob routing. Optional Playwright MCP + Caveman plugin in
Claude Code skill (and standalone Bash CLI) for managing Cloudflare Tunnel ingress routes. No browser, no dashboard. Persistent HTTPS subdomains for any local port.
Agent skill for using the Yali AI GPT-image2 inspiration library, templates, image generation/editing paths, and PPT workflow routing.
Pluggable MCP tool modules that give agent workflows their capabilities — GitHub, GitLab, Sentry, Slack, Jira, Notion, memory, datasets, and more.
AutoChatPlus Chat Widget - A customizable floating action button that opens a chat interface, allowing users to interact with an AI assistant.
Lightweight webhook monitoring SDK — know when your webhooks break
Pi package that gives agents a change_reasoning tool for adjusting reasoning level when substantial work is likely.
Unified LLM API with automatic model discovery and provider configuration
MERV Playwright client: Merv-Local JSON/HTML reports, MervPlaywrightHandler plugin steps, optional screenshots — aligned with Java merv-client.
Production-grade environment variable management with runtime validation for JavaScript and TypeScript. Safely validate, parse, and manage environment variables across runtimes with dotenv compatibility and modern framework support.
Session-aware routing layer for AI-assisted software delivery workflows
Implementation-first AI spec knowledge base skill for Claude Code
A lightweight, human-readable dialogue scripting system for games. Write conversations in plain JSON — no compiler required.