Skip to main content

AI Integration

To enable AI copy the ai configuration file to the local directory and set the enabled flag to true.

enabled: true
ollamaUrl: http://localhost:11434
openAiKey: xxxx
prompts:
  translate:
    model: ollama/yi
    template: |
      Translate the following text to {{language}}:
      
      {{content}}
  resource:
    model: ollama/codellama
    template: |
      Do you see problems in the following kubernetes resource?
      
      {{content}}

You can optimize the prompt templates for the models you use. The {{content}} placeholder will be replaced with the content of the current klubernetes resource. The {{language}} placeholder will be replaced with the language requested language.

You can use the following models:

  • ollama/*: all models from the ollama service, you have to install the models manually before. Set ollamaUrl to the correct url.
  • openai/*: all models from the openai service, you have to set the openAiKey in the configuration file.

You can add options for the model like this ollama/codellama/temperature=0.3&timeout=60.

  • temperature: The temperature of the model, a higher temperature will result in more creative answers.
  • timeout: The timeout in seconds for the model.