example.commitronrc

# Commitron configuration file
# This file should be placed at ~/.commitronrc

# AI provider configuration
ai:
  # Available providers: openai, gemini, ollama, claude, custom
  provider: ollama
  # Your API key for the selected provider
  api_key: your-api-key-here
  # The model to use - varies by provider
  model: mistral:latest
  # Control creativity (0.0-1.0): lower values are more deterministic, higher values more creative
  temperature: 0.7
  # Set to true to see detailed debugging information about AI requests and responses
  debug: false
  # Maximum tokens in AI response (increase for longer commit messages or more complex changes)
  max_tokens: 1000
  # Optional custom system prompt - overrides default AI instructions
  # Leave empty to use the default prompt that matches the selected convention
  # For conventional commits, a default prompt like this will be used:
  # system_prompt: |
  #   Generate a concise git commit message written in present tense for the following code changes.
  #   Commit message must be a maximum of 72 characters.
  #   Include a brief descriptive body explaining the changes.
  #   Choose an appropriate type from: feat, fix, docs, style, refactor, perf, test, build, ci, chore, revert.
  #system_prompt: ""
  # Only needed for Ollama, defaults to http://localhost:11434
  ollama_host: http://localhost:11434
  # Only needed for custom provider - URL of your local GPT API endpoint
  # Examples:
  # - Open WebUI: http://localhost:8080/v1/chat/completions
  # - Local API server: http://localhost:8000/v1/chat/completions
  # - Custom endpoint: https://your-api.example.com/v1/chat/completions
  #api_endpoint: http://localhost:8080/v1/chat/completions

# Commit message configuration
commit:
  # Available conventions: none, conventional, custom
  convention: conventional
  # Whether to include a message body after the subject
  include_body: true
  # Maximum length for the subject line
  max_length: 72
  # Maximum length for the commit body - keep this concise to avoid truncation
  max_body_length: 400
  # Only used when convention is 'custom'
  # custom_template: "{{type}}({{scope}}): {{subject}}"

# Context settings for AI
context:
  # Whether to include file names in the context
  # Provides basic information about which files changed
  include_file_names: true

  # Whether to include the git diff in the context
  # This is the most important setting for accurate commit messages
  # STRONGLY RECOMMENDED to keep this enabled
  include_diff: true

  # Maximum length for the context to avoid exceeding token limits
  # Increase for large changes, decrease if you hit token limits
  # For more accurate messages with large diffs, increase this value
  max_context_length: 4000

  # Include statistics about file changes (+/- lines)
  # Helps AI understand the magnitude and type of changes
  include_file_stats: true

  # Include brief description of what each file does (based on first few lines)
  # Helps AI understand the purpose of each file
  include_file_summaries: true

  # Show the first N lines of each file for additional context
  # Set to 0 to disable
  # Note: This is skipped when include_diff is true to avoid duplication
  show_first_lines_of_file: 5

  # Include high-level repository structure for better context
  # Helps for changes that affect multiple parts of the codebase
  # May not be needed for simple changes
  include_repo_structure: false

# User interface configuration
ui:
  # Enable TUI (Text User Interface) for better visualization
  # Shows staged files and commit message with confirmation
  enable_tui: true

  # Ask for confirmation before finalizing the commit message
  confirm_commit: true

  # Maximum number of files to display in the TUI (0 = no limit)
  # Useful for commits with many files
  display_files_limit: 20