logo

Git Connectors

The Git Connectors enable seamless integration with Git repositories, allowing you to index and search through your codebase, documentation, and other version-controlled content.

Git Loader Architecture

Git Connectors Architecture

1.1 Git Repository Loader

Load and process content from Git repositories with support for multiple branches, commit history, and file types.

1.1.1 Use Cases

  • Code documentation indexing
  • Technical documentation search
  • Source code analysis
  • Version history tracking

1.1.2 Required Configuration

  • Repository Path: Local path where the repository will be cloned
  • Clone URL: HTTPS or SSH repository URL
  • Branch: Specific branch to clone (defaults to main/master)
  • File Filter: Glob patterns to include specific files (e.g., "*.md", "*.py")
  • Content Filter: Regular expressions to filter file content

1.1.3 Output Data

  • File content with metadata
  • File path and name
  • Last commit information
  • Branch information

1.2 GitDigest Loader

A specialized loader for processing GitHub repositories with focus on commit history and repository analytics.

1.2.1 Required Configuration

  • GitHub Repository: Full repository name (owner/repo)
  • Authentication: GitHub Personal Access Token (if required)

1.2.2 Output Data

  • Commit history and messages
  • Repository statistics
  • Contributor information
  • Issue and pull request metadata

Note: Ensure proper Git credentials are configured and repository access is granted before using the connector.

Tip: Use .gitignore patterns to exclude unnecessary files and optimize loading performance.