Browse agent skills

Switch Claude Code between AI model providers in one command. Supports DeepSeek, MiniMax, Groq, Moonshot, OpenRouter, and more.

Generate CLAUDE.md and AGENTS.md context files for AI coding assistants from an interactive CLI.

NotebookLM-centered research orchestration pipeline

Opinionated agentic development lifecycle orchestrator

AI chat plugin for Puck editor — drop-in replacement using Groq/Llama with Unsplash image support

Claude Code skill — deep-dive analysis during planning. Traces execution paths, finds related code, surfaces edge cases.

Utilitários de ancoragem OCR para extração de biomarcadores de PDFs de resultados laboratoriais

One source of truth for your AI rules. Sync to Claude, Cursor, Copilot, Windsurf and more.

Reliability layer for AI outputs: schema validation, repair, retry, fallback, and budget guards

Drive GitHub PRs to merge-ready: watch CI, fix failures, resolve review comments.

WeChat AI Bot — Bridge WeChat to Claude, Qwen, DeepSeek and more

AI layer for Moteur: providers, credits, translation, generation

An SSH runner plugin for the xyOps workflow automation system.

Amodal agent runtime core

A meta-prompting, context engineering and spec-driven development system for Claude Code.

Reusable AI context and installable skills for coding agents, broadly installable via skills.sh.

CLI-first Gemini web automation via CDP — AI chat, image generation, session management with machine-safe JSON output

Abstract skill interfaces for AI — delivery, testing, architecture, security, and more. Skills define WHAT; AI decides HOW.

Auto-generate 'Everything [Team] shipped in X days' release calendar posters from real X posts. A Claude Code skill.

Agentic context compiler for AI-assisted development in TypeScript/JavaScript projects

用于管理和快速切换 Claude Code 服务商配置的命令行工具。支持内置 Anthropic 服务商（含代理注入）和任意自定义服务商。

The AI research agent for the terminal

CLI multi-agent refactoring orchestrator powered by Pi Coding Agent and OpenRouter

Skills-Native Application runtime — server, providers, session management, database, and CLI

OpenCode plugin: multiple GitHub Copilot accounts

No description available

TypeScript port of the Semantic-Similarity Rating (SSR) methodology for converting LLM responses to Likert-scale probability distributions

Generate /ai endpoint specs for your service — make it AI-readable in minutes

One .ai/ folder to rule all AI coding tool configs

AI agent skill implementing Google's TurboQuant compression algorithm (ICLR 2026) — 6x KV cache memory reduction, 8x speedup, zero accuracy loss. Compatible with Claude Code, Codex CLI, and all Agent Skills-compatible tools.