← 全部工具

agent-team-foundation/first-tree

热度 65 更新于 AI 与 Agent

First-tree routes work to the right agent, gives it the same context your team has, and loops humans in only when the rules say so. Lives in your GitHub. Open source.

githubauto-collected

安装

暂未验证可直接使用的安装命令,请查看项目官方文档或 Release。

<p align="center" <picture <source media="(prefers-color-scheme: dark)" srcset="assets/banner-dark.png" <source media="(prefers-color-scheme: light)" srcset="assets/banner-light.png" <img src="assets/banner-light.png" alt="First Tree" width="100%" </picture </p

<p align="center" <a href="https://first-tree.ai/?utmsource=github&utmmedium=readme&utmcampaign=nav-app"<strongOpen App</strong</a &middot; <a href="#get-started"<strongGet Started</strong</a &middot; <a href="#how-it-works"<strongHow It Works</strong</a &middot; <a href="docs/quickstart.md"<strongQuickstart</strong</a &middot; <a href="https://github.com/agent-team-foundation/first-tree/discussions"<strongDiscussions</strong</a </p

<p align="center" <a href="https://www.npmjs.com/package/first-tree"<img src="https://img.shields.io/npm/v/first-tree?style=for-the-badge&color=FFD700&label=npm" alt="npm version"</a <a href="https://github.com/agent-team-foundation/first-tree/actions/workflows/ci.yml"<img src="https://img.shields.io/github/actions/workflow/status/agent-team-foundation/first-tree/ci.yml?style=for-the-badge&label=CI" alt="CI"</a <a href="https://github.com/agent-team-foundation/first-tree/blob/main/LICENSE"<img src="https://img.shields.io/badge/License-Apache%202.0-green?style=for-the-badge" alt="License: Apache 2.0"</a <a href="https://github.com/agent-team-foundation/first-tree/stargazers"<img src="https://img.shields.io/github/stars/agent-team-foundation/first-tree?style=for-the-badge&color=blueviolet" alt="GitHub stars"</a </p

<p align="center" English | <a href="READMEzh-CN.md"中文</a </p

Try first-tree 🌳 free — the fastest way to give every agent your team's shared context.

First-Tree

Context-grounded agentic work for teams.

First Tree is an open-source workspace where AI agents work from your team's shared context, not isolated prompts.

At the center is Context Tree: a team-maintained memory of decisions, ownership, repos, responsibilities, constraints, and prior work. Agents read it before they work; useful outcomes can flow back into it after the work is done.

The result is a human-agent work loop where every task can start with more team context, and every useful outcome can make the next task smarter.

<p align="center" <img src="assets/workspace-screenshot.png" alt="First Tree workspace: an agent reporting back on a GitHub issue, with team context and participants alongside" width="100%" </p

<p align="center" <sub<bThe work loop in practice.</b An agent reports back on what it shipped for a GitHub issue &mdash; while the linked issue, the team's chat history, and every human and agent participant stay in view, so the next task starts from the same shared context.</sub </p

<div align="center" <table <tr <td align="center"<strongWorks<br/with</strong</td <td align="center"<picture<source media="(prefers-color-scheme: dark)" srcset="assets/logos/claude-code-dark.svg"<img src="assets/logos/claude-code-light.svg" width="32" alt="Claude Code" /</picture<br/<subClaude Code</sub</td <td align="center"<picture<source media="(prefers-color-scheme: dark)" srcset="assets/logos/codex-dark.svg"<img src="assets/logos/codex-light.svg" width="32" alt="Codex" /</picture<br/<subCodex</sub</td <td align="center"<picture<source media="(prefers-color-scheme: dark)" srcset="assets/logos/github-dark.svg"<img src="assets/logos/github-light.svg" width="32" alt="GitHub" /</picture<br/<subGitHub</sub</td <td align="center"<picture<source media="(prefers-color-scheme: dark)" srcset="assets/logos/mcp-dark.svg"<img src="assets/logos/mcp-light.svg" width="32" alt="MCP" /</picture<br/<subMCP</sub</td </tr </table </div

---

Why First Tree

There are many AI agent workspaces. First Tree's core difference is the context loop:

user intent -> read team context -> context-aware agent work
-> human review/control -> durable outcome -> updated team context

This loop solves the part that usually breaks when teams start using agents: