Search

Safe, read-only database query connector with guard-rail policy enforcement. Built on @narai/connector-toolkit. Ships a JSON-envelope CLI and a library API. Supports PostgreSQL, MySQL, SQLite, SQL Server, MongoDB, DynamoDB, and Oracle via pluggable driver

Zero-friction shared memory for multi-agent AI systems. Pass URL pointers instead of token-expensive blobs.

AgentID integration for LangChain — identity, discovery, and trust for AI agents

AgentID integration for CrewAI — identity and trust for crew agents

Synthetic user research and LLM eval harness — research-grade rigor for developers who can't afford a research team.

Sustainable Foraging environment for multi-agent RL

AgentID integration for AutoGen — identity and trust for conversable agents

OpenAI-compatible Chat Completions provider for Zeno (OpenAI, OpenRouter, LiteLLM, Azure v1, Ollama, vLLM).

Guard your LLM agents in CI. Snapshot tests that catch behavioral regressions when models, prompts, or vendors change.

Run coding agents (Claude Code, Codex, Aider, OpenHands, …) against prompts in container sandboxes via Harbor

AI Memory Operating System — Graph-RAG, temporal truth maintenance, actionable schemas, selective encryption, sub-200ms hybrid retrieval.

Scan an AI-agent repo and emit runtime-supervisor wrappers, policies, and combo playbooks.

Drop-in Anthropic SDK wrapper for AgentLoop — adds memory retrieval and turn logging to messages.create calls.

Drop-in OpenAI SDK wrapper for AgentLoop — adds memory retrieval and turn logging to chat.completions.create calls.

Claude Agent SDK provider for Zeno (Anthropic Claude with sub-agents, built-in tools, and SDK hooks).

Production-grade PostgreSQL MCP Server. Empowers LLMs with 40+ tools for deep database interaction, including schema inspection, secure query execution, performance analysis, and administrative tasks. Built for reliable AI-driven development.

AI Agent Governance Framework — Not to make AI "write better", but to make AI "not mess up"

Agentic AI role-playing game framework — narration, NPCs, lore, and character creation are autonomous plugin agents. Build interactive fiction with DeepSeek / OpenAI / Anthropic / Qwen.

A Model Context Protocol (MCP) server that provides tools for executing shell commands in the background and retrieving their results asynchronously. This server is designed for AI agents that need to run long-running or blocking commands without blocking the main thread.

🐱 PurrCat的核心仓库