Verified project record

HKUDS/AutoAgent

hkuds/autoagent generates and orchestrates LLM agents, tools, and workflows using natural language without requiring users to write code. The project is locally runnable, requires external LLM API keys, and includes a ready-to-use deep research mode.

Project overview

Project type: AI Agent · Workflow
Use cases: Search & Research · Automation
Deployment: Refer to project documentation
License: MIT

Best for

Developers and general users who want to build custom agents and workflows using natural language without writing code.
Researchers who need a ready-to-use assistant for deep information retrieval and report generation.

Key capabilities

Automatically constructs and orchestrates collaborative agent systems purely through natural dialogue.
Allows anyone, regardless of coding experience, to create and customize their own agents, tools, and workflows using natural language alone.
Dynamically creates, optimizes and adapts agent workflows based on high-level task descriptions.
Enables controlled code generation for creating tools, agents, and workflows through iterative self-improvement, supporting both single agent creation and multi-agent workflow generation.
Enables controlled code generation for creating tools, agents, and workflows through iterative self-improvement, supporting both single agent creation and multi-agent workflow generation.
Ready-to-use multi-agent system accessible through user mode on the start page serving as a comprehensive AI research assistant designed for information retrieval, complex analytical tasks, and comprehensive report generation.
Allows creating tools, agents, and workflows using natural language alone without workflow.
Allows creating agent workflows using natural language description.

Limitations and risks

The workflow editor mode does not support tool creation temporarily.
The system requires configuring and routing requests to external LLM APIs such as OpenAI API, Anthropic API, DeepSeek API, Gemini API, Hugging Face API, Groq API, xAI API, OpenRouter API, and Mistral API, incurring associated paid service costs.
GPU requirements, minimum hardware specifications, and telemetry data collection practices are not documented.

Getting started

Install the Python package via pip install -e ., configure API keys in a .env file, and run the auto main command. Docker is used for environment containerization.

Evidence and sources

GitHub project description: "AutoAgent: Fully-Automated and Zero-Code LLM Agent Framework"
README: AutoAgent is a **Fully-Automated** and highly **Self-Developing** framework that enables users to create and deploy LLM agents through **Natural Language Alone**.
README: Automatically constructs and orchestrates collaborative agent systems purely through natural dialogue, eliminating the need for manual coding or technical configuration.
README: More features coming soon! 🚀 **Web GUI interface** under development.
README: Dynamically creates, optimizes and adapts agent workflows based on high-level task descriptions, even when users cannot fully specify implementation details.

Reviewed at: 2026-07-18T03:44:14Z