Cloving Config Command Documentation

Learn how to use the cloving config command to manage your Cloving project configuration.

Usage

cloving config

Description

The cloving config command is used to configure AI models for your Cloving project. It allows you to select AI providers, models, set API keys, and configure various settings like primary usage, priority, review, and trust settings for sensitive information.

Options

This command does not have specific options. It uses an interactive prompt for configuration.

Interactive Prompts

The cloving config command uses a series of interactive prompts to guide you through the configuration process:

  • Select a provider: Choose an AI model provider from the available options.
  • Select a category: Choose a category of models from the selected provider.
  • Select an AI model: Choose a specific AI model from the selected category.
  • Enter your API key: Provide the API key for the selected AI model.
  • Set as primary: Choose whether to set this model as the primary model.
  • Enter the priority: Set the priority for this model (0-100).
  • Review prompts: Choose whether to review prompts before they are sent to the model.
  • Trust with sensitive information: Indicate whether you trust the model with sensitive information.
  • Configure another model: Choose whether to configure another model.

Examples

Example Configuration Session

Here is an example of what a configuration session might look like:

? Select a provider: openai
? Select a category: gpt
? Select an AI model you'd like to use: openai:gpt:4o
? Enter your API key for openai:gpt:4o: YOUR_API_KEY_HERE
? Set this model as the primary? Primary models are used by default if your prompt fits within its context window. You can configure backup models for larger prompts. (Y/n) Y
? Enter the priority for this model (0-100). Higher priority AI APIs are chosen as long as the prompt fits its context window: 100
? Do you want to review all prompts before they are sent to openai:gpt:4o? (Y/n) Y
? Do you trust openai:gpt:4o with sensitive information? (y/N) N
? Do you want to configure another model? (Y/n) N

This session configures the OpenAI GPT-4o model with the highest priority, sets it as the primary model, enables prompt review, and does not trust it with sensitive information.

Sample Configuration File

Below is an example of what the ~/.clovingconfig file might look like after configuration:

globalSilent=false

[models.claude.claude-3-5:sonnet-20240620]
apiKey=
primary=true
priority=100
silent=false
trust=true

[models.openai.gpt:4o]
apiKey=
primary=false
priority=50
silent=false
trust=false

[models.ollama.llama3:70b-instruct]
apiKey=
primary=false
priority=null
silent=true
trust=true

[models.gemini.gemini-1.5:pro-latest]
apiKey=
primary=false
priority=10
silent=false
trust=false

[models.mistral.mistral:large-latest]
apiKey=
primary=false
priority=5
silent=false
trust=false

[models.mistral.open-mixtral:8x22b]
apiKey=
primary=false
priority=4
silent=false
trust=false