Skip to content

Security: intertwine/vertex-mcp-chatbot

Security

SECURITY.md

Security Policy

Supported Versions

Version Supported
0.1.x

Reporting a Vulnerability

If you discover a security vulnerability in this project, please follow these steps:

  1. DO NOT open a public issue
  2. Email the details to the project maintainer (see CODE_OF_CONDUCT.md for contact)
  3. Include:
    • Description of the vulnerability
    • Steps to reproduce
    • Potential impact
    • Suggested fix (if any)

Security Considerations

When using this chatbot with MCP servers:

Authentication

  • Store sensitive credentials in environment variables, not in code
  • Use .env files for local development (never commit these)
  • OAuth tokens are stored in .mcp_tokens/ - ensure this directory is gitignored

MCP Server Security

  • Always validate input in MCP servers
  • Implement path restrictions for filesystem operations
  • Use authentication for production MCP servers
  • Limit tool permissions to what's necessary

Network Security

  • Use HTTPS for HTTP/SSE transports in production
  • Verify SSL certificates
  • Be cautious with self-signed certificates

Example Server Notes

The example servers in examples/mcp-servers/ are for demonstration only:

  • The filesystem server has basic path validation but should be enhanced for production
  • The weather server uses mock data and is safe
  • The OAuth servers use test credentials - replace for production use

Best Practices

  1. Environment Variables: Use ${VAR_NAME} syntax in MCP config for secrets
  2. File Permissions: Restrict access to .mcp_tokens/ and .env files
  3. Input Validation: Always validate and sanitize user input in MCP servers
  4. Least Privilege: Grant minimal necessary permissions to MCP tools
  5. Audit Logging: Log MCP tool usage for security monitoring

There aren't any published security advisories