Multi-platform AI agent framework — deploy specification-driven agents to GitHub Copilot, Cursor, Claude Code, Windsurf, and the Open Plugins standard in one command
Neo Skills: A Universal AI Agent Skills Extension
A simple, extensible AI agent built on marco-harness.
TENET — The operating system for AI agent teams
DIDComm Message Receipts protocol implementation for @credo-ts/core
Multiplayer CLI
Agent-first MCP server for WeChat Mini Program debugging, automation, and lightweight testing
Command-line interface for the Duvo public API
Agentic visual regression for BlazeDiff. Auto-discovers routes, captures deterministic screenshots, runs CI checks.
TypeScript definitions for s3-zip
Context Code es un asistente de desarrollo para la terminal. Puede revisar tu proyecto, editar archivos, ejecutar comandos y apoyarte en tareas reales de programacion.
Constructive CLI SDK - Auto-generated GraphQL CLI with ORM client, context management, and interactive prompts
Cross-platform AI agent skill converter - Write once, run on any agent
Module to create a basic Pino LoggerConfig to support Google Cloud structured logging
Compozy CLI
Git workflow automation for coding agents and MCP clients
Spaghetti I/O Processing, Interpolation, Correlation and beyond - now with plugins
Pi-Pompom — A 3D raymarched virtual pet with voice, ambient weather sounds, AI side chat, agent tracking, and 23 layered sound effects for Pi CLI
LangChain.js adapters for Model Context Protocol (MCP)
Sleek term-based dashboard for AI coding CLI quotas
1. Create, at the root of your project, a file called `.env.local`. 2. Open the file and declare the following variable to replace the default urls: ```text VITE_NETWORK_OVERRIDES={"ai":"http://localhost:8080","agent-tools":"http://localhost:8082", "genAi
Shared **guard runtime** for Praxis shell policy: argv parsing (`shell-quote`), `evaluateShellProposal` / `evaluateArgv` / `gateShellCommand`, JSONL audit append, and the **shell approval bridge** under `.praxis/guard/bridge`.
Supercharge Claude Code with semantic code intelligence. 94% fewer tool calls • 77% faster exploration • 100% local.
JWT-authenticated credential proxy for upstream LLM providers