Jaco-AskChatGPT
Ask arbitrary questions to ChatGPT.
Uses reverse engineered python-api as interface.
Setup:
Requires an OpenAI-Account.
Follow the V1-setup-guide to get your authentication data.
Save it in the config.yaml file afterwards.
Debugging:
docker build -t skill_jaco_askchatgpt_amd64 - < skills/skills/Jaco-AskChatGPT/image_amd64.dockerfile
# Run without display
docker run --network host --rm \
--volume `pwd`/skills/skills/Jaco-AskChatGPT/:/Jaco-Master/skills/skills/Jaco-AskChatGPT/:ro \
--volume `pwd`/skills/skills/Jaco-AskChatGPT/skilldata/:/Jaco-Master/skills/skills/Jaco-AskChatGPT/skilldata/ \
--volume `pwd`/userdata/config/:/Jaco-Master/userdata/config/:ro \
-it skill_jaco_askchatgpt_amd64
# Run with real display
xhost + && docker run --network host --rm \
--volume `pwd`/skills/skills/Jaco-AskChatGPT/:/Jaco-Master/skills/skills/Jaco-AskChatGPT/:ro \
--volume `pwd`/skills/skills/Jaco-AskChatGPT/skilldata/:/Jaco-Master/skills/skills/Jaco-AskChatGPT/skilldata/ \
--volume `pwd`/userdata/config/:/Jaco-Master/userdata/config/:ro \
--volume /tmp/.X11-unix:/tmp/.X11-unix \
--env DISPLAY --env QT_X11_NO_MITSHM=1 \
--privileged \
-it skill_jaco_askchatgpt_amd64
# Create a virtual display in the docker container
export DISPLAY=:123
Xvfb $DISPLAY -screen $DISPLAY 1280x1024x16 &
# Try loading a webpage only
python3 /Jaco-Master/skills/skills/Jaco-AskChatGPT/try_chrome.py
# Try an example question
python3 /Jaco-Master/skills/skills/Jaco-AskChatGPT/try_api.py
# Start the skill's action
python3 /Jaco-Master/skills/skills/Jaco-AskChatGPT/action_answer.py