mirror of
https://github.com/aljazceru/gpt-engineer.git
synced 2026-02-06 21:56:12 +01:00
34513f651b6b4e04f014d6022f5e05e2553b6a89
GPT Engineer
Specify what you want it to build, the AI asks for clarification, and then builds it. Even if it is complex.
Philosophy
The philosophy of this project is that it should be be
- Simple to get value
- Flexible and simple to add new own "AI steps" (see
steps.py) - Incrementally build towards a user experience of:
- high level prompting
- giving feedback to the AI that it will remember over time
- Fast handovers back and forth between AI and human
- No databases, all computation is "resumable" and persisted to filesystems
Usage
Install:
pip install -r requirements.txt
Run:
- Create a new empty folder with a
main_promptfile (or copy the example foldercp example -r my-new-project) - Fill in the
main_promptin your new folder - run
python main.py my-new-project
Results:
- Check the generated files in my-new-project/workspace_clarified
Features
You can specify the "identity" of the AI agent by editing the files in the identity folder.
Editing the identity, and evolving the main_prompt, is currently how you make the agent remember things between projects.
Each step in steps.py will have its communication history with GPT4 stored in the logs folder, and can be rerun with scripts/rerun_edited_message_logs.py.
Languages
Python
93.5%
Makefile
6.5%