Skip to main content

Configuration Guide

This guide explains how to configure CodeCompass for optimal performance with Claude 3.7 Sonnet and other language models.

Configuration File​

CodeCompass uses a JSON configuration file located at ~/.codecompass/config.json by default. You can specify a different location using the --config flag when running commands.

Basic Configuration​

A minimal configuration file looks like this:

{
  "api_keys": {
    "anthropic": "your_anthropic_api_key"
  },
  "default_model": "anthropic/claude-3-sonnet"
}

Full Configuration Options​

Here's a complete configuration file with all available options:

{
  "api_keys": {
    "anthropic": "your_anthropic_api_key",
    "openai": "your_openai_api_key",
    "mistral": "your_mistral_api_key",
    "cohere": "your_cohere_api_key"
  },
  "default_model": "anthropic/claude-3-sonnet",
  "max_tokens": 4096,
  "temperature": 0.7,
  "chunk_size": 8000,
  "chunk_overlap": 200,
  "models": {
    "fast": "anthropic/claude-3-haiku",
    "balanced": "anthropic/claude-3-sonnet",
    "powerful": "anthropic/claude-3-opus",
    "cost_effective": "mistral/mistral-large-latest"
  },
  "cache": {
    "enabled": true,
    "directory": "~/.codecompass/cache",
    "max_size_mb": 1024,
    "ttl_hours": 24
  },
  "semantic_chunking": {
    "enabled": true,
    "language_parsers": {
      ".py": "python",
      ".js": "javascript",
      ".ts": "typescript",
      ".java": "java",
      ".cpp": "cpp",
      ".go": "go",
      ".rb": "ruby",
      ".rs": "rust"
    }
  },
  "web_interface": {
    "port": 3000,
    "host": "127.0.0.1",
    "enable_cors": false
  },
  "logging": {
    "level": "info",
    "file": "~/.codecompass/logs/codecompass.log"
  }
}

Configuration Options Explained​

API Keys​

"api_keys": {
  "anthropic": "your_anthropic_api_key",
  "openai": "your_openai_api_key",
  "mistral": "your_mistral_api_key"
}

Set your API keys for different providers. The Anthropic key is required for Claude 3.7 Sonnet, while others are optional if you want to use multiple models through LiteLLM.

Model Selection​

"default_model": "anthropic/claude-3-sonnet",
"models": {
  "fast": "anthropic/claude-3-haiku",
  "balanced": "anthropic/claude-3-sonnet",
  "powerful": "anthropic/claude-3-opus",
  "cost_effective": "mistral/mistral-large-latest"
}
  • default_model: The model to use when none is specified
  • models: Named aliases for different models, making it easier to switch between them

Generation Parameters​

"max_tokens": 4096,
"temperature": 0.7
  • max_tokens: Maximum number of tokens to generate in responses
  • temperature: Controls randomness in the output (0.0 = deterministic, 1.0 = creative)

Semantic Chunking​

"chunk_size": 8000,
"chunk_overlap": 200,
"semantic_chunking": {
  "enabled": true,
  "language_parsers": {
    ".py": "python",
    ".js": "javascript"
    // ... other language mappings
  }
}
  • chunk_size: Maximum size of each chunk in characters
  • chunk_overlap: Number of characters to overlap between chunks
  • semantic_chunking: Configuration for language-aware chunking:
    • enabled: Whether to use semantic chunking (true) or simple chunking (false)
    • language_parsers: Mapping of file extensions to language parsers

Caching​

"cache": {
  "enabled": true,
  "directory": "~/.codecompass/cache",
  "max_size_mb": 1024,
  "ttl_hours": 24
}
  • enabled: Whether to cache API responses
  • directory: Where to store cache files
  • max_size_mb: Maximum cache size in megabytes
  • ttl_hours: Time-to-live for cache entries in hours

Web Interface​

"web_interface": {
  "port": 3000,
  "host": "127.0.0.1",
  "enable_cors": false
}
  • port: The port to run the web interface on
  • host: The host address to bind to
  • enable_cors: Whether to enable Cross-Origin Resource Sharing

Logging​

"logging": {
  "level": "info",
  "file": "~/.codecompass/logs/codecompass.log"
}
  • level: Log level (debug, info, warning, error)
  • file: Path to log file

Environment Variables​

You can also configure CodeCompass using environment variables, which will override the config file values:

# API Keys
export CODECOMPASS_ANTHROPIC_API_KEY=your_anthropic_api_key
export CODECOMPASS_OPENAI_API_KEY=your_openai_api_key
export CODECOMPASS_MISTRAL_API_KEY=your_mistral_api_key

# General Configuration
export CODECOMPASS_DEFAULT_MODEL=anthropic/claude-3-sonnet
export CODECOMPASS_MAX_TOKENS=4096
export CODECOMPASS_TEMPERATURE=0.7

# Chunking Configuration
export CODECOMPASS_CHUNK_SIZE=8000
export CODECOMPASS_CHUNK_OVERLAP=200
export CODECOMPASS_ENABLE_SEMANTIC_CHUNKING=true

# Caching
export CODECOMPASS_CACHE_ENABLED=true
export CODECOMPASS_CACHE_DIR=~/.codecompass/cache

# Web Interface
export CODECOMPASS_WEB_PORT=3000
export CODECOMPASS_WEB_HOST=127.0.0.1

Command-Line Overrides​

Many configuration options can be overridden on the command line:

codecompass process --model anthropic/claude-3-opus \
                   --max-tokens 8192 \
                   --temperature 0.5 \
                   --no-cache \
                   path/to/file.py

Configuration Profiles​

You can create multiple configuration profiles for different use cases:

# Create a new profile
codecompass config create-profile large-context

# Edit a profile
codecompass config edit-profile large-context --max-tokens 8192 --chunk-size 16000

# Use a profile
codecompass --profile large-context process path/to/file.py

Next Steps​

Now that you've configured CodeCompass, you can: