Browse agent skills

向上天下一张恋爱定单 · 一个有仪式感的 Claude Code Skill:访谈式描绘理想伴侣画像,生成可发送给宇宙的《致宇宙的定单》

Deterministic verification rules for AI-generated API integrations

AI-powered development workflow orchestrator for Claude Code

Thin HTTP client for the VellumCharter entitlements + billing API

The agent framework where nothing is done until a grader returns a verdict — verification + grounded perception (AgentVision eyes). Reserved; under active design.

Living, plain-English documentation for AI-built projects. Almanac auto-creates and continuously syncs a hierarchy of project docs through any coding agent (Claude Code, Codex, Gemini) and any git host.

Parse TypeScript source into a knowledge graph for autonomous code optimization

Scaffold a Claudius AI chat widget project in one command.

GOCR — small, CPU OCR: document -> text + bounding boxes as JSON. ~38 MB, no GPU. Node port of the g-ocr Python package.

Bizar opencode plugin — loop detection, status reporting, handoff signal, background agents, and slash commands + visual plan flow for subagent activity

Instrumentation for Cost-to-Code attribution.

n8n community node for the Sequence Platform v1 API (accounts, transfers, rules, transactions)

Database-agnostic, inline-edit headless CMS engine for React / Next.js apps

Portable agent skill for structured System Design mentoring (Cursor, Claude Code, Codex)

Self-hosted backend for the Handsfree iOS app — talk to Claude Code or Codex hands-free from your phone.

Nulz Code — free, bring-your-own-key agentic coding CLI. Multi-provider (Gemini, Claude, OpenRouter, OpenAI, Ollama + more), tools, scraping, automation.

Complete Discord Bot integration for n8n with all features - DM, Mentions, Server Management, Roles, Threads, Reactions, Voice, Advanced Tools and more

支語警察 — 偵測並改正中文裡的中國用語(支語)、改回台灣慣用語。純 Python 核心,可獨立執行或接任何 AI Agent(Claude Code/Codex/Cursor/Gemini)。四級判定+三級模式+人工回饋閉環。

Kōji - Hybrid SQL + Recursive + Vector Database

Kōji - Hybrid SQL + Recursive + Vector Database

Corum is a Git-native design graph for service architecture. It models components (APIs, domain models, schemas, fields) as nodes with typed edges, and exposes the graph through MCP tools so AI assistants can reason about your architecture.

pi coding-agent extension that runs every read/write/edit/bash op inside a per-session Docker container

Verity CLI — run your Playwright suite (fast/full tier), detect flakes via reruns, and report to Verity.

Static security scanner for AI agent skill packages. Detects malicious SKILL.md files and bundled scripts before they run.

AgentOS terminal CLI — drive a real LLM agent on the microkernel; strongest-UX shell, advanced agent core

Loopful — a Claude Code skill that enforces a disciplined, verifiable engineering workflow for multi-step development tasks.

Installable Agent Skills for GitHub Copilot, Claude Code, and Cursor — copywriting (F6S & LinkedIn) and NVIDIA Inception tooling.

OpenClaw Inworld speech plugin.

OpenClaw Gradium speech plugin.

Unified LLM API with automatic model discovery and provider configuration