经 AI Skill Hub 精选评估,TaskForce AI代理 获评「强烈推荐」。这款Agent工作流在功能完整性、社区活跃度和易用性方面表现出色,AI 评分 8.0 分,适合有一定技术背景的用户使用。
TaskForce AI代理 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排,将复杂的多步骤任务拆解为清晰的自动化流程,实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成,适合构建数据处理管线、业务自动化和 AI 辅助决策系统。
TaskForce AI代理 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排,将复杂的多步骤任务拆解为清晰的自动化流程,实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成,适合构建数据处理管线、业务自动化和 AI 辅助决策系统。
# 方式一:npm 全局安装 npm install -g taskforce-aiagent # 方式二:npx 直接运行(无需安装) npx taskforce-aiagent --help # 方式三:项目依赖安装 npm install taskforce-aiagent # 方式四:从源码运行 git clone https://github.com/taskforce-aiagent/taskforce-aiagent cd taskforce-aiagent npm install npm start
# 命令行使用
taskforce-aiagent --help
# 基本用法
taskforce-aiagent [options] <input>
# Node.js 代码中使用
const taskforce_aiagent = require('taskforce-aiagent');
const result = await taskforce_aiagent.run(options);
console.log(result);
# taskforce-aiagent 配置说明 # 查看配置选项 taskforce-aiagent --config-example > config.yml # 常见配置项 # output_dir: ./output # log_level: info # workers: 4 # 环境变量(覆盖配置文件) export TASKFORCE_AIAGENT_CONFIG="/path/to/config.yml"
<p align="center"> <a href="https://github.com/taskforce-aiagent/taskforce-aiagent"> <img src="https://img.shields.io/github/stars/taskforce-aiagent/taskforce-aiagent?style=social" alt="GitHub Stars"> </a> <a href="https://www.npmjs.com/package/taskforce-aiagent"> <img src="https://img.shields.io/npm/v/taskforce-aiagent" alt="npm version"> </a> <a href="https://img.shields.io/npm/dm/taskforce-aiagent"> <img src="https://img.shields.io/npm/dm/taskforce-aiagent" alt="downloads"> </a>
<a href="https://github.com/taskforce-aiagent/taskforce-aiagent/blob/main/LICENSE"> <img src="https://img.shields.io/github/license/taskforce-aiagent/taskforce-aiagent" alt="license"> </a> <a href="https://codecov.io/gh/taskforce-aiagent/taskforce-aiagent/branch/main/graph/badge.svg"> <img src="https://img.shields.io/github/actions/workflow/status/taskforce-aiagent/taskforce-aiagent/main.yml?branch=main" alt="build status"> </a> <a href="https://codecov.io/gh/taskforce-aiagent/taskforce-aiagent"> <img src="https://codecov.io/gh/taskforce-aiagent/taskforce-aiagent/branch/main/graph/badge.svg" alt="coverage"> </a> <a href="https://taskforce-aiagent.github.io/taskforce-aiagent/"> <img src="https://img.shields.io/badge/docs-latest-blue" alt="docs"> </a> <br/ > <a href="https://github.com/marcoaras"> <img src="https://img.shields.io/badge/creator-marcoaras-blue?logo=github" alt="creator"> </a> <a href="https://github.com/jeudomos"> <img src="https://img.shields.io/badge/creator-jeudomos-white?logo=github" alt="creator"> </a> <a href="https://github.com/pribrahimsari"> <img src="https://img.shields.io/badge/creator-pribrahimsari-red?logo=github" alt="creator"> </a> </p>
TaskForce is a modular, open, and production-ready TypeScript framework for orchestrating LLM-powered autonomous AI agents, task pipelines, dynamic toolchains, RAG workflows, and memory/retrieval systems.
It is designed for real-world use-cases such as document analysis, automation, agent collaboration, enterprise search, and more.
Docs: Agents ・ Tasks ・ Tools ・ Built-in Tools Memory & Retrieval ・ Fine-tuning ・ TaskForce Core
---
---
| Component | Description |
|---|---|
| **Agent** | An LLM-powered persona with a role, tools, memory, and goals. ([details](./docs/en/agent.md)) |
| **Task** | A unit of work, assigned to an agent, with input/output and dependencies. ([details](./docs/en/task.md)) |
| **Tool** | A function, API, or skill agents can invoke (search, code, web, RAG, vision, etc). ([details](./docs/en/tool.md)) |
| **Memory** | Vector memory, chat history, and RAG—per-agent or global. ([details](./docs/en/memory.md)) |
| **TaskForce** | The orchestrator managing agents, tasks, tools, execution, delegation, and coordination. ([details](./docs/en/taskForce.md)) |
---
---
npm install taskforce-agent
---
import {
TaskForce,
Agent,
Task,
DirectoryReadTool,
BraveSearchTool,
ExecutionMode,
} from "taskforce-agent";
// 1. Define agents
const researcher = new Agent({
id: "researcher",
name: "Market Researcher",
goal: "Find up-to-date insights on AI trends",
tools: [new BraveSearchTool(), new DirectoryReadTool()],
memory: true,
});
const writer = new Agent({
id: "writer",
name: "Writer",
goal: "Write a summary blog post based on research",
tools: [],
});
// 2. Define tasks
const researchTask = new Task({
id: "t1",
description: "Research the latest AI trends",
expected_output: "A list of 3 trending developments in AI",
agent: "researcher",
});
const writeTask = new Task({
id: "t2",
description: "Write a markdown blog post summarizing research",
expected_output: "A 4-paragraph markdown blog post",
agent: "writer",
inputFromTask: "t1",
});
// 3. Create TaskForce
const tf = new TaskForce({
agents: [researcher, writer],
tasks: [researchTask, writeTask],
executionMode: ExecutionMode.AiDriven, // or "Hierarchical", "Parallel"
verbose: true,
});
// 4. Run the workflow
await tf.run({ topic: "AI in 2025" });
---
Create a .env file in your project root to enable all tool and memory features:
```dotenv
OPENAI_API_KEY="your-openai-key" DEEPSEEK_API_KEY="your-deepseek-key" ANTHROPIC_API_KEY="your-anthropic-key" GEMINI_API_KEY="your-gemini-key" MISTRAL_API_KEY="your-mistral-key" BRAVE_API_KEY="your-brave-key" SERPAPI_API_KEY="your-serpapi-key"
高质量的开源AI工作流框架
AI Skill Hub 为第三方内容聚合平台,本页面信息基于公开数据整理,不对工具功能和质量作任何法律背书。
建议在沙箱或测试环境中充分验证后,再部署至生产环境,并做好必要的安全评估。
✅ MIT 协议 — 最宽松的开源协议之一,可自由商用、修改、分发,仅需保留版权声明。
AI Skill Hub 点评:TaskForce AI代理 的核心功能完整,质量优秀。对于自动化工程师和运维人员来说,这是一个值得纳入个人工具库的选择。建议先在非生产环境试用,再逐步推广。
| 原始名称 | taskforce-aiagent |
| Topics | agentaitypescript |
| GitHub | https://github.com/taskforce-aiagent/taskforce-aiagent |
| License | MIT |
| 语言 | TypeScript |
收录时间:2026-06-29 · 更新时间:2026-06-29 · License:MIT · AI Skill Hub 不对第三方内容的准确性作法律背书。
选择 Agent 类型,复制安装指令后粘贴到对应客户端