Skip to content

Security: awesome-archive/bunny-agent

Security

SECURITY.md

Security Policy

Reporting a Vulnerability

If you discover a security vulnerability in Bunny Agent, please report it privately.

Do NOT open a public issue.

How to Report

Send an email to: security@vikadata.com

Include the following information:

  • Description of the vulnerability
  • Steps to reproduce
  • Potential impact
  • Any proof-of-concept code (if available)

What Happens Next

  1. You will receive an acknowledgment within 48 hours
  2. We will investigate the vulnerability
  3. We will work on a fix
  4. You will be notified when the fix is released
  5. We will credit you in the release notes (if desired)

Supported Versions

Version Support Status
Latest release ✅ Supported
Previous releases ⚠️ Best effort
Development branch ⚠️ Best effort

Security fixes are backported to the latest supported version.


Security Best Practices

When using Bunny Agent:

API Keys

  • Never commit API keys to version control
  • Use environment variables for sensitive data
  • Rotate API keys regularly
  • Use different keys for development and production

Sandbox Configuration

  • Always use sandboxes in production environments
  • Limit sandbox permissions to minimum required
  • Regularly update sandbox images
  • Monitor sandbox resource usage

Running Bunny Agent

  • Keep dependencies updated: pnpm update
  • Review dependencies for known vulnerabilities: npm audit
  • Use official packages from npm

Production Deployment

  • Enable HTTPS for all endpoints
  • Implement rate limiting on API endpoints
  • Validate and sanitize all user inputs
  • Log security-relevant events for monitoring

Known Security Considerations

Sandbox Isolation

Bunny Agent uses sandboxed environments for code execution:

  • Local mode: Runs on your local filesystem with your permissions
  • Cloud mode: Runs in isolated containers (E2B, Sandock, Daytona)

For production use, always prefer cloud sandboxes for isolation.

Agent Outputs

  • Review agent-generated code before deployment
  • Validate file paths and operations
  • Implement output filtering for sensitive data

File Access

Agents may access files within their working directory:

  • Set appropriate working directory boundaries
  • Audit templates for file access patterns
  • Use volume mounts with care in cloud sandboxes

Security Updates

We will announce security updates through:

  • GitHub Security Advisories
  • Release notes for affected versions
  • Email to security@vikadata.com subscribers

Contact

For security-related questions:

There aren't any published security advisories