本地桌面端执行型AI助手 - Open Interpreter 风格 | Local Desktop Execution-Oriented AI Assistant
Framework-agnostic Verifiable Service Agent SDK
Pixel-art RPG companion for Codex. Watch Codex reason, run commands, edit files, and quest through beautiful biomes as you code together.
Interactive TUI for managing installed skills for AI coding agents (Claude Code, Codex, OpenClaw, and more)
Resolves debug namespace like magic.
Prints a error string or an error object to console.error
A simple, extensible AI agent built on marco-harness.
AI-powered code review automation for GitLab/GitHub using Claude Code
General-purpose agent with transport abstraction, state management, and attachment support
TypeScript SDK for Auggie
Use Environment Variables in String
简约的插件化 Agent 框架
Wasm/JavaScript bindings for Subduction (browser and Node.js)
Evidence-backed ROADMAP.md generator and sync tool for AI coding agents.
codefresh logs
npm package for rari
OpenCode plugin — orchestrator-driven skills with 9 expert sub-agents, Phase Gate pipeline, and SQLite context storage
<!-- ~ Copyright (c) 2025-2026 Datalayer, Inc. ~ ~ BSD 3-Clause License -->
| A library for building Web5 identity management applications | | ------------------------------------------------------------ |
Pi extension that keeps working on a /goal until the agent marks it complete.
Invoke ECMAScript scripts in any command line JS engine.
Pi package for reliable web access with explicit search, fetch, and headless boundaries.
Shared HTTP utilities for agent-safe action and workflow runner adapters.
Learning system for Claude Code — avoids repeating mistakes across sessions