pip install -r requirements.txt
Download RCE from their website and set it up.
Edit config.cfg by adding following information:
Parameter | Description |
---|---|
openai_api_key |
Path to RCE |
ollama_server_ip |
OpenAI API Key (if you are using ChatGPT models) |
rce_path |
IP to the Ollama server (if you are using local LLMs like Llama) |
Following Paramters can be changed in the config.cfg file:
Parameter | Description | Example |
---|---|---|
controller_model |
the LLM that performs actions on the GUI of RCE | gpt-4o-2024-08-06 |
evaluator_model |
the LLM that evaluates the GUI for Problems | gpt-4o-2024-08-06 |
icon_description_model |
LLM that gives a textual description of the icons on the GUI | gpt-4o-2024-08-06 |
max_saved_actions |
maximum number of actions that are saved | 20 |
max_actions |
maximum number of actions that are being performed | 200 |
step_by_step |
if true, requires user input for performing any number of actions | true |
test_only_tool_integration |
if true, only tests the Tool Integration Wizard of RCE (For testing all of RCE not all capabilities for the controler Agent are implemented) | true |
After running automated_gui_testing.py the results can be found in the ./temp/
folder.
Please see the file LICENSE.md for further information about how the content is licensed.