GitHub

Small library to build agents which are controlled by large language models (LLMs) which is heavily inspired by langchain .
The goal was to get a better grasp of how such an agent works and understand it all in very few lines of code.
Langchain is great, but it already has a few more files and abstraction layers, so I thought it would be nice to build the most important parts of a simple agent from scratch.
Some more infos are in this Hacker News discussion from April 5th 2023 and the related blog post .
How it works
The agent works like this:
* It gets instructed by a prompt which tells it the basic way to solve a task using tools
* Tools are custom build components which the agent can use
* So far, I've implemented the ability to execute Python code in a REPL, to use the Google search and to search on Hacker News
* The agent runs in a loop of Thought, Action, Observation, Thought, ...
* The Thought and Action (with the Action Input to the action) are the parts which are generated by an LLM
* The Observation is generated by using a tool (for example the print outputs of Python or the text result of a Google search)
* The LLM gets the new information appended to the prompt in each loop cycle and thus can act on that information
* Once the agent has enough information it provides the final answer
For more details on how it works, check out this blog post
How to use it
You can install this library locally by running:
inside it's directory after cloning it.
You also need to provide the following env variables:
* OPENAI_API_KEY to use the OpenAI API (obtainable at: https://platform.openai.com/account/api-keys)
* SERPAPI_API_KEY to use the Google Search in case you use that tool (obtainable at: https://serpapi.com/)
You can simply export them in bash like: export OPENAI_API_KEY='sh-lsdf....'
Then you can run the script python run_agent.py and ask your question.