GeneralAgent  by CosmosShadow

Python agent framework for LLM integration

created 1 year ago
453 stars

Top 67.6% on sourcepulse

GitHubView on GitHub
Project Summary

GeneralAgent is a Python-native agent framework designed to seamlessly integrate Large Language Models (LLMs) with Python capabilities. It empowers users to build sophisticated agents for complex business processes, enabling tool invocation via Python code interpretation rather than LLM function calls, robust serialization for state persistence, and efficient task execution through self-calling and stack memory to minimize LLM interactions.

How It Works

The framework's core innovation lies in its Python code interpreter, which allows agents to directly call tools and execute Python code, bypassing the limitations of LLM-native function calling. This approach enhances reliability and flexibility. GeneralAgent also features a unique "self-call" mechanism and "stack memory" to reduce LLM token usage by intelligently managing task decomposition and execution, making it efficient for complex, multi-step operations.

Quick Start & Requirements

  • Install via pip: pip install GeneralAgent
  • Requires an OpenAI API key (or compatible endpoint) configured via .env file or directly in code.
  • Official documentation and examples are available.

Highlighted Details

  • Tool Calling: Leverages Python code interpreter for tool invocation, not LLM function calls.
  • Serialization: Supports saving and resuming agent memory and Python execution state.
  • Multi-modal Input: Accepts text and image inputs for richer interactions.
  • Model Agnostic: Designed to work with various LLMs by overriding skills.llm_inference or using OpenAI SDK compatibility.

Maintenance & Community

  • The project is actively maintained by CosmosShadow.
  • Community engagement is encouraged via WeChat group (QR code in README).

Licensing & Compatibility

  • The README does not explicitly state a license. Compatibility for commercial use or closed-source linking is undetermined.

Limitations & Caveats

  • The absence of a specified license poses a significant adoption blocker for commercial or closed-source projects.
  • AI Search functionality requires configuring a SERPER_API_KEY and installing selenium.
Health Check
Last commit

8 months ago

Responsiveness

1 day

Pull Requests (30d)
0
Issues (30d)
0
Star History
4 stars in the last 90 days

Explore Similar Projects

Starred by Chip Huyen Chip Huyen(Author of AI Engineering, Designing Machine Learning Systems) and Robert Stojnic Robert Stojnic(Creator of Papers with Code).

Agent-S by simular-ai

1.2%
6k
Agentic framework for autonomous computer interaction
created 9 months ago
updated 18 hours ago
Feedback? Help us improve.