The bot connect to an ollama endpoint and queries said endpoint whenever it gets mentioned
[Describe the installation process, including any dependencies that the user needs to install before running your bot. If your bot is available on PyPI, include a link to the PyPI page.]
In order to run the bot you need the following settings:
- .env file that contains the following info :
- DISCORD_TOKEN -> The discord token for the bot to be running
- level -> log level, so far the only accepted level is debug
- OLLAMA_URL -> The ollama server url with the following format
http://ollama_host:ollama_port
- A .log directory
It is also needed to run pip3 install -r deps/requirements.txt
Another way to have the bot running could be by running the docker-compose file docker-compose up -d
Make sure you have a .env file in the current directory with the following structure
DISCORD_TOKEN = YOUR_DISCORD_TOKEN
level = debug
OLLAMA_URL = http://ollama_host:ollama_port
OLLAMA_MODEL = llm_model
To check if the container is running docker-compose ps
It would be good to work on a branch called dev
When the work on the branch is done commit it to dev
The gh workflow will automatically create a PR
In order to interact with the bot you just have to mention it in your message @bot_name text