ATaC MCP Server
ATaC-team/ATaC
[](https://agentrank-ai.com/tool/ATaC-team--ATaC/?utm_source=badge&utm_medium=readme&utm_campaign=agentrank_badge) Are you the maintainer of ATaC-team/ATaC? Claim this listing →
Record and Replay AI Agent Trajectories.
Add AgentRank to Claude Code Discover and compare tools like ATaC-team/ATaC — your AI finds the right one automatically
Get API Access → claude mcp add agentrank -- npx -y agentrank-mcp-server Overview
ATaC-team/ATaC is a TypeScript tool licensed under MIT. Record and Replay AI Agent Trajectories. Topics: agent-infra, agents, mcp, skills.
Ranked #11168 out of 24846 indexed tools.
Actively maintained with commits in the last week.
Ecosystem
agent-infraagentsmcpskills
Score Breakdown
Stars 15% 303
303 stars → growing community
Freshness 25% 7d ago
Last commit 7d ago → actively maintained
Issue Health 25% 50%
No issues filed → no history to score
Contributors 10% 2
2 contributors → solo project
Dependents 25% 0
No dependents → no downstream usage
npm Downloads N/A
PyPI Downloads 13% 1/wk
1 weekly installs → early adoption
Forks 11
Description Brief
License MIT
Weights: Freshness 20% · Issue Health 20% · Dependents 22% · Stars 10% · Contributors 8% · How we score →
How to Improve
Description low impact
Dependents medium impact
Matched Queries
From the README
<div align="center">
[English](#english) | [中文](#中文)
</div>
## 中文
ATaC是一个面向 Agent 工作流的工具运行时,帮助 Agent 通过编排图代码复用资产。
它主要提供三类能力:
- 统一工具注册:把本地 Agent 工具和 MCP 工具纳入同一个运行时
- Agent 自编排 graph:让 Agent 生成的 graph 代码可以直接复用这些已注册能力
- 一键运行:加载并执行这些 graph 工作流
把工具注册到 ATaC,让 Agent 用代码把这些工具组织成流程,并逐步沉淀为可复用、可维护的 graph 工作流。
### 🤖 最小接入示例
#### 1. 注册 Agent 内置工具和 MCP 工具
```python
import asyncio
import inspect
import sys
from pathlib import Path
from atac import AtacService
import atac.subprocess as subprocess
from langchain_core.tools import tool
from langchain_mcp_adapters.client import MultiServerMCPClient
service = AtacService()
@tool
def bash(command: str) -> str:
"""在当前运行目录里执行 shell 命令。"""
completed = subprocess.run(
command,
shell=True,
capture_output=True,
text=True,
)
if completed.returncode != 0:
raise RuntimeError(completed.stderr.strip() or "bash failed")
return completed.stdout.rstrip("\n")
service.register_langgraph_tools([bash] Read full README on GitHub → Get the weekly AgentRank digest
Top movers, new tools, ecosystem insights — straight to your inbox.