Skip to content

Vinothsuku/agentLite

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

18 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

agentLite - Tiny Agentic Toolkit

A lightweight and modular toolkit for prototyping AI agents that can perform tasks like web searching, content summarization, and feedback analysis. Designed for developers who want to experiment with intelligent agents without the overhead of a full-fledged framework.

Key Features:

  • Web Search Integration: Searches the web using DuckDuckGo and extracts relevant results.
  • Content Summarization: Uses OpenAI's GPT-4 (by default but its configurable) to generate concise summaries of retrieved content
  • Feedback-Driven Adaptation: Agents can analyze user feedback and adapt their behavior over time.
  • Memory and Logging: Agents maintain a memory of actions and results, with detailed logging for debugging and analysis.
  • Modular Design: Easily add new actions, modify existing ones, or integrate with other APIs.

Use Cases:

  • Automating research tasks.
  • Building conversational AI assistants.
  • Prototyping intelligent agents for specific workflows.

Get Started:

  • Clone the repository.
  • Install dependencies: pip install openai requests beautifulsoup4 faiss-cpu numpy.
  • Set your OpenAI API key in the config.yaml file and run the agent: python main.py --task "Your task description" --config "config.yaml"

Contribute: Feel free to fork, star, or contribute to this project! Suggestions, bug reports, and feature requests are welcome

Blog post link

About

A Simple Modular Agentic Framework

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages