This document outlines the security measures implemented in LLMCal and best practices for secure usage.

• No Hardcoded Credentials: All API keys and sensitive data have been removed from source code
• Environment Variable Support: Secure loading from .env files and system environment variables
• PopClip Integration: Secure credential handling through PopClip's encrypted options system
• Credential Validation: Format validation and strength checking for API keys

• Log Sanitization: Automatic removal of sensitive data from all log entries
• API Request/Response Logging: Safe logging with credential redaction
• Structured Logging: Multiple log levels (DEBUG, INFO, WARN, ERROR, CRITICAL)
• Log Rotation: Automatic log file rotation with size limits
• Restricted Permissions: Log files created with 600 permissions (owner read/write only)

• Temporary Files: Secure creation and cleanup of temporary files
• File Permissions: Automatic setting of restrictive permissions on sensitive files
• Secure Cleanup: Overwrite temporary files with random data before deletion

• Format Validation: Validation of API key formats and credential strength
• Dependency Checking: Verification of required credentials based on features used
• Connectivity Testing: Optional API connectivity validation

// Config.json - INSECURE
{
  "zoom_account_id": {
    "default": "COoBTtIEQ-ynFT_zpUL6jw"  // ❌ Hardcoded credential
  }
}

# test_cases.sh - INSECURE
export POPCLIP_OPTION_ANTHROPIC_API_KEY="sk-ant-api03-37F5..." # ❌ Hardcoded API key

# calendar.sh - INSECURE
log "API 响应: $RESPONSE"  # ❌ Logs sensitive API responses

// Config.json - SECURE
{
  "zoom_account_id": {
    "description": "Enter your Zoom Account ID",
    "secure": true  // ✅ No default values, marked as secure
  }
}

# test_cases.sh - SECURE
if [ -z "$ANTHROPIC_API_KEY" ]; then
    log "Error: Missing ANTHROPIC_API_KEY environment variable"  # ✅ Environment variable required
    exit 1
fi

# calendar.sh - SECURE
log_api_response "$http_code" "" "$(sanitize_json_response "$response")"  # ✅ Sanitized logging

# Run the security setup wizard
./setup_security.sh

1. Create Environment File

   cp .env.example .env
   chmod 600 .env

2. Configure API Keys

   # Edit .env file
   nano .env
   
   # Add your credentials:
   ANTHROPIC_API_KEY=your_actual_api_key_here
   ZOOM_ACCOUNT_ID=your_zoom_account_id
   ZOOM_CLIENT_ID=your_zoom_client_id
   ZOOM_CLIENT_SECRET=your_zoom_client_secret

3. Validate Configuration

   source LLMCal.popclipext/lib/config_validator.sh
   run_validation true

• ANTHROPIC_API_KEY: Your Anthropic Claude API key

• ZOOM_ACCOUNT_ID: Your Zoom account ID
• ZOOM_CLIENT_ID: Your Zoom app client ID
• ZOOM_CLIENT_SECRET: Your Zoom app client secret
• ZOOM_EMAIL: Email associated with Zoom account
• ZOOM_NAME: Display name for Zoom meetings

• LOG_LEVEL: Logging level (DEBUG, INFO, WARN, ERROR, CRITICAL) - default: INFO
• MAX_LOG_SIZE: Maximum log file size in bytes - default: 10MB
• MAX_LOG_FILES: Number of rotated log files to keep - default: 5

• Use environment variables or PopClip secure options for credentials
• Set restrictive file permissions (600) on sensitive files
• Regularly rotate API keys
• Monitor log files for security events
• Keep the extension updated
• Use the provided security validation tools

• Hardcode credentials in source files
• Commit .env files to version control
• Share log files containing sensitive data
• Use weak or predictable API keys
• Run the extension with elevated privileges unnecessarily

The security logging system automatically:

• Redacts API keys, tokens, and credentials from all log entries
• Sanitizes email addresses (partial redaction)
• Removes long alphanumeric strings that might be credentials
• Logs security events with appropriate severity levels

• DEBUG: Detailed information for troubleshooting
• INFO: General information about operations
• WARN: Warning conditions that should be noted
• ERROR: Error conditions that need attention
• CRITICAL: Critical issues that require immediate action

The system automatically validates:

• API key format correctness
• Credential strength and complexity
• Required vs optional credential dependencies
• File permission security
• Configuration completeness

# Run comprehensive security check
source LLMCal.popclipext/lib/config_validator.sh
run_validation true

If you suspect a security issue:

1. Immediate Actions

   • Rotate all affected API keys immediately
   • Check log files for suspicious activity
   • Review recent extension usage

2. Investigation

   • Check git history for accidentally committed credentials
   • Scan all files for hardcoded secrets
   • Review access logs from API providers

3. Prevention

   • Update to latest version with security fixes
   • Re-run security validation
   • Review and update security practices

• v2.0: Comprehensive security overhaul
  • Removed all hardcoded credentials
  • Implemented secure logging with sanitization
  • Added configuration validation
  • Added secure file handling
  • Created security utilities library

• Monitor for security updates
• Review CHANGELOG.md for security-related changes
• Re-run security validation after updates

For security-related questions or to report security issues:

1. Check this documentation first
2. Run the built-in security validation tools
3. Review log files for error messages
4. Open an issue in the project repository (for non-sensitive issues)

Before using LLMCal in production:

• Removed all hardcoded credentials from source code
• Created .env file with proper permissions (600)
• Configured all required API keys
• Ran security validation successfully
• Verified logging sanitization is working
• Added .env to .gitignore
• Set up log rotation and monitoring
• Reviewed file permissions on all extension files
• Tested API connectivity with validation tools
• Documented credential rotation procedures

Note: Security is an ongoing process. Regularly review and update your security practices, credentials, and keep the extension updated with the latest security improvements.