Configuration

Configure drep with config.yaml to customize platform adapters, LLM settings, rate limiting, caching, and more.

Configuration File

drep automatically discovers your configuration file in this order:

1. Explicit path: --config /path/to/config.yaml (if provided)
2. Environment variable: DREP_CONFIG environment variable
3. Project-specific: ./config.yaml (current directory)
4. User config: Platform-dependent:
   • macOS: ~/Library/Application Support/drep/config.yaml
   • Linux: ~/.config/drep/config.yaml

This means you can run drep scan owner/repo without specifying --config - drep will automatically find your configuration!

drep init

The sections below explain all available configuration options if you need to customize beyond the defaults.

Platform Configuration

Configure at least one platform adapter (Gitea, GitHub, or GitLab). You can enable multiple platforms simultaneously.

Gitea

gitea:
  url: http://localhost:3000
  token: ${GITEA_TOKEN}
  repositories:
    - owner/*  # All repos for this owner
    - org/specific-repo  # Specific repository

GitHub

github:
  token: ${GITHUB_TOKEN}
  repositories:
    - owner/repo
    - organization/*
  url: https://api.github.com  # Or GitHub Enterprise URL

GitLab

gitlab:
  url: https://gitlab.com  # Or your self-hosted GitLab URL
  token: ${GITLAB_TOKEN}
  repositories:
    - owner/*  # All projects for this owner
    - org/specific-project  # Specific project

LLM Settings

Configure your LLM backend: local models (LM Studio, Ollama, llama.cpp, vLLM) or AWS Bedrock (Claude 4.5 with ZDR).

Local LLM Configuration

For local backends (LM Studio, Ollama, llama.cpp):

llm:
  enabled: true
  endpoint: http://localhost:1234/v1  # LM Studio
  model: qwen3-30b-a3b
  temperature: 0.2
  timeout: 120
  max_tokens: 8000
  max_retries: 3
  retry_delay: 2
  max_concurrent_global: 5
  max_concurrent_per_repo: 3
  requests_per_minute: 60
  max_tokens_per_minute: 80000
  cache:
    enabled: true
    directory: ~/.cache/drep/llm
    ttl_days: 30
    max_size_gb: 10

LLM Options Explained

AWS Bedrock Configuration

For AWS Bedrock (Claude 4.5 with Zero Data Retention):

llm:
  enabled: true
  provider: bedrock  # Required for Bedrock

  bedrock:
    region: us-east-1
    model: anthropic.claude-sonnet-4-5-20250929-v1:0

  # General LLM settings
  temperature: 0.2
  max_tokens: 4000

  # Rate limiting (lower for Bedrock)
  max_concurrent_global: 3
  requests_per_minute: 30

  # Cache configuration
  cache:
    enabled: true
    ttl_days: 30

Supported Bedrock Models:

• anthropic.claude-sonnet-4-5-20250929-v1:0 - Claude Sonnet 4.5 (recommended)
• anthropic.claude-haiku-4-5-20251001-v1:0 - Claude Haiku 4.5 (fast & cost-effective)

Enterprise Benefits:

• Zero Data Retention - Your data is never stored or used for training
• Data Sovereignty - Data stays in your AWS region and account
• Compliance - Meets HIPAA, GDPR, SOC 2 requirements

Cache Settings

drep caches LLM responses to improve performance and reduce costs. Typical cache hit rates are 80%+.

Documentation Settings

documentation:
  enabled: true
  custom_dictionary:
    - drep
    - Gitea
    - PyPI
  markdown_checks: false  # Optional: lightweight linting

custom_dictionary: Add project-specific terms to skip spell-checking (e.g., product names, technical jargon).

markdown_checks: Enable non-LLM markdown linting (trailing whitespace, broken links, malformed headings). Use drep lint-docs for standalone linting.

Database Configuration

drep uses SQLite by default for tracking scans and caching.

database_url: sqlite:///./drep.db

For production deployments, you can use PostgreSQL or other SQLAlchemy-supported databases:

database_url: postgresql://user:password@localhost/drep

Environment Variables

Use environment variables for sensitive values like API tokens:

# Platform tokens (recommended over hardcoding)
export GITEA_TOKEN="your-token"
export GITHUB_TOKEN="your-token"
export GITLAB_TOKEN="your-token"

# Override config file location
export DREP_CONFIG="/path/to/config.yaml"

# Override LLM endpoint
export DREP_LLM_ENDPOINT="http://localhost:11434"

Complete Configuration Example

Here's a fully annotated config.yaml with all available options:

# Platform configuration
gitea:
  url: http://localhost:3000
  token: ${GITEA_TOKEN}
  repositories:
    - your-org/*

github:
  token: ${GITHUB_TOKEN}
  repositories:
    - owner/repo
    - organization/*
  url: https://api.github.com

# Documentation settings
documentation:
  enabled: true
  custom_dictionary:
    - drep
    - Gitea
    - PyPI
  markdown_checks: false

# Database
database_url: sqlite:///./drep.db

# LLM configuration
llm:
  enabled: true
  endpoint: http://localhost:1234/v1
  model: qwen3-30b-a3b
  temperature: 0.2
  timeout: 120
  max_tokens: 8000
  max_retries: 3
  retry_delay: 2
  max_concurrent_global: 5
  max_concurrent_per_repo: 3
  requests_per_minute: 60
  max_tokens_per_minute: 80000
  cache:
    enabled: true
    directory: ~/.cache/drep/llm
    ttl_days: 30
    max_size_gb: 10

Next Steps

• Set up your LLM backend with model recommendations and troubleshooting
• View examples of webhook setup, CI/CD integration, and Docker deployment
• Explore the CLI for all available commands and options