Configuration Guide¶

Rusty Buddy provides a flexible configuration system using the config.toml file located in your project's .rusty directory. This guide outlines the configuration options available to tailor Rusty Buddy to your preferences.

Configuration File Location¶

The config.toml file is located within the .rusty directory in your project's root folder:

your_project/
├── .rusty/
│   └── config.toml

Note: Rusty Buddy will automatically search for .rusty/config.toml starting from your current directory and moving up through parent directories. Make sure you are working in or beneath a configured project directory.

If the .rusty directory or config.toml does not exist, you can generate it by running:

rusty-buddy init

Configurable Options¶

1. Default Persona¶

Set the default persona that Rusty Buddy uses if none is specified during a chat session.

default_persona = "rust"

2. AI Models Configuration¶

Configure the AI models for different functionalities:

[ai]
chat_model = "openai_complex"
commit_model = "openai_fast"
wish_model = "openai_complex"
chat_timeout_secs = 30  # Timeout duration in seconds

chat_model: Model used for chat sessions.
commit_model: Model used for generating commit messages.
wish_model: Model used for the wish command.
chat_timeout_secs: (New) Timeout duration for AI responses in seconds.

Example:

[ai]
chat_model = "openai_complex"
commit_model = "openai_fast"
wish_model = "openai_complex"
chat_timeout_secs = 90  # Set timeout to 15 minutes

3. Models Definition¶

Define the AI models and their backends:

[[models]]
name = "openai_fast"
api_name = "gpt-4o-mini"
backend = "OpenAI"

[[models]]
name = "openai_complex"
api_name = "gpt-4o-2024-08-06"
backend = "OpenAI"

4. Personas Configuration¶

Customize or add new personas to enhance your AI interactions:

[[personas]]
name = "rust"
chat_prompt = "As a seasoned Rust developer..."
file_types = ["rs", "toml", "md", "yml"]

5. Log Level Settings¶

Control logging outputs via your .rusty/config.toml:

console_log_level = "Warn"  # Controls what appears in your terminal
file_log_level = "Info"     # Controls what appears in 'rusty.log'

Set to Debug, Info, Warn, or Error as needed.

Configuring the Timeout Duration¶

New Feature: You can now configure the timeout duration for AI responses using the chat_timeout_secs parameter in your config.toml file. This setting ensures that all chat AI interactions timeout consistently based on your preference.

Setting the Timeout¶

Open your config.toml file located in .rusty/config.toml.
Under the [ai] section, add or modify the chat_timeout_secs parameter:

[ai]
chat_model = "your_chat_model"
commit_model = "your_commit_model"
wish_model = "your_wish_model"
chat_timeout_secs = 60  # Timeout set to 1 minute

Save the file. The new timeout duration will be applied to all subsequent AI interactions.

Understanding Timeout Behavior¶

The chat_timeout_secs parameter defines the maximum duration (in seconds) Rusty Buddy will wait for a response from the AI backend before timing out.
Adjusting this value can help manage long-running requests or prevent delays in your workflow.

Example:

Shorter Timeout (e.g., chat_timeout_secs = 30): Useful for quick interactions where responses are expected promptly.
Longer Timeout (e.g., chat_timeout_secs = 90): Allows the AI more time to process complex or resource-intensive requests.

Advanced Configuration¶

Feel free to explore and adjust other settings in the config.toml file to better suit your development needs. Always ensure that the syntax is correct to prevent configuration errors.

Note: Any changes to the config.toml file require restarting Rusty Buddy to take effect.

Troubleshooting¶

If you encounter issues:

Invalid Configuration Syntax: Ensure your config.toml follows proper TOML syntax.
Timeout Errors: If requests are timing out too soon or taking too long, adjust the chat_timeout_secs value accordingly.
Contact Support: If problems persist, reach out via Contact Support.

Happy Coding with Rusty Buddy!