MCP Agent Graph - Python SDK for Agent Development Framework
Wellknown found it in public sources; nobody has proven control of it yet. Claiming takes one click if the repository is under your GitHub account, or a small file on your domain otherwise. Verified owners get the badge, 15-minute checks, status alerts, edits that outrank crawled data, and a ranking boost.
Agents can do it too: POST https://wellknown.network/api/v1/claims with {"agent":"mcp-agent-graph","method":"well_known_file"} — machine-readable steps at claim.json, guide at /docs/claim.
Everything here was measured by our prober or read from a registry. Nothing is self-reported.
Attributed to the source that supplied each field. Treated as claims, not facts.
 English | [中文](README_CN.md) > Preface: mcp-agent-graph is an efficient, lightweight, and easy-to-use Agent development framework. As an Agent developer, you might use visual workflow orchestration tools like Dify or Coze; you might prefer code development and have experience with langgraph, crewai, Google ADK, etc. How is mcp-agent-graph different from these frameworks? What makes it worth recommending? This article will introduce the project's design philosophy and functional features. We aim to help you understand this project in the shortest time possible - perhaps you'll find it perfectly matches your needs. > Core Project Philosophy: need -> agent 📚 [Documentation](https://keta1930.github.io/mcp-agent-graph/#) | 📦 [PyPI Package](https://pypi.org/project/mcp-agent-graph/) | 📄 [Design Philosophy, Features & Roadmap](docs/一文说清%20mcp-agent-graph%20设计理念、功能特点、未来规划.pdf) ## 📚 Table of Contents - [🚀 Deployment Guide](#-deployment-guide) - [Frontend Deployment](#frontend-deployment) - [Backend Deployment](#backend-deployment) - [Option 1: PyPI Installation (Recommended)](#option-1-pypi-installation-recommended) - [Option 2: Using Conda](#option-2-using-conda) - [Option 3: Using uv (Recommended)](#option-3-using-uv-recommended) - [Quick Start](#quick-start) - [✨ Core Features](#-core-features) - [1️⃣ From need to Agent](#1️⃣-from-need-to-agent) - [2️⃣ AI-Generated MCP Tools (From need to MCP)](#2️⃣-ai-generated-mcp-tools-from-need-to-mcp) - [3️⃣ Nested Graphs (Hierarchical World)](#3️⃣-nested-graphs-hierarchical-world) - [4️⃣ Graph to MCP Server](#4️⃣-graph-to-mcp-server) - [5️⃣ Visual Graph Editor](#5️⃣-visual-graph-editor) - [6️⃣ Node as Agent](#6️⃣-node-as-agent) - [7️⃣ Agent Trading and Transfer](#7️⃣-agent-trading-and-transfer) - [8️⃣ Python SDK Deep Integration](#8️⃣-python-sdk-deep-integration) - [📝 Summary](#-summary) - [🖼️ Frontend Feature Showcase](#️-frontend-feature-showcase) …
Mapped onto the structured taxonomy from declared text and observed tool names. Confidence shown for derived entries.
Every source is kept verbatim. Field changes are logged as events.