Search

🤖 AI Agent Framework Guide (中文) | Scion · AutoGen · CrewAI · LangGraph · MetaGPT · Dify · Coze — 深度对比 + 选型决策树 | Chinese developer guide for choosing the right multi-agent framework

MD-driven agent CLI for code, infrastructure, and automation. 6 providers, 8 tools, zero SDK dependencies.

Self-hosted shared memory layer for AI coding agents like Codex, Claude Code, Cursor, and MCP-compatible tools.

Halo · structural facts about your code, exposed to any AI agent over MCP. Long-lived daemon parses your repo with tree-sitter, builds a call graph + git-aware index, and gives Cursor / Claude Code / Cody / Cline / Continue / Windsurf / Zed ten tools (callersOf, calleesOf, pathBetween, recentlyChangedNear, rca, …) for grounded code reasoning.

Manage AI agent memory with a filesystem-native daemon that consolidates, extracts, and organizes markdown memories through pluggable LLMs

Quality-gated granularity control for multi-agent LLM pipelines

An experimental attempt at self-evolution for the agent team

Local-first desktop app: 16 AI agents from 8 LLM providers debate any topic — with live argument mapping, fact-check badges, and conflict detection.

Official SDKs and CLI for the Berry Communication Protocol — agent-facing API of the V-Box community platform.

An agentic AI travel planner application that generates and refines travel itineraries according to user input, optimising plans based on constraints such as budget, must-visit places, and travel preferences, while maintaining conversational context to iteratively improve and adapt the trip over time.

Production-ready multi-agent CLI with unified memory

The AI Harness for Python Developers — wrap any LLM with budget control, persistent memory, sandboxed execution, multi-agent orchestration, and production observability.

Python SDK for StacyVM — self-hosted microVM sandboxes for LLMs

SuperAGI integration for xLink identity-based authentication (autonomous AI agents, 603× faster M2M auth)

Identity-based M2M authentication for LangChain agents (Python bindings)

AutoGen integration for xLink identity-based authentication (multi-agent agentic coordination, 603× faster M2M auth)

Haystack integration for xLink identity-based authentication (RAG pipelines, 603× faster M2M auth)

A fuse box for your AI bill. Drop-in spend protection and runaway-loop circuit breaker for Claude and OpenAI agents. Includes Python and Node SDKs plus a CLI.

Core runtime for [MAST](https://github.com/andreban/mast-ai), a TypeScript library that runs AI agent loops in the browser. The "think-act" loop, tool registry, and conversation state all live client-side; remote backends serve only as stateless reasoning

Block AI coding agents (Claude Code, Cursor, Codex, Aider) on the same quality gates your humans hit at git commit.