Skip to content

Security: hyperpolymath/echidnabot

SECURITY.md

Security Policy

Supported Versions

The following versions of echidnabot are currently supported with security updates:

Version Supported
0.1.x

Reporting a Vulnerability

We take security seriously. If you discover a security vulnerability in echidnabot, please report it responsibly:

How to Report

  1. Email: Send details to security@hyperpolymath.org
  2. Encryption: Use our GPG key at https://hyperpolymath.org/gpg/security.asc
  3. Do NOT open a public GitHub issue for security vulnerabilities

What to Include

  • Description of the vulnerability
  • Steps to reproduce
  • Potential impact assessment
  • Suggested fix (if any)

Response Timeline

  • Initial acknowledgment: Within 48 hours
  • Status update: Within 7 days
  • Resolution target: Within 30 days for critical issues

What to Expect

  • If the vulnerability is accepted, we will:

    • Work on a fix and coordinate disclosure timing with you
    • Credit you in the security advisory (unless you prefer anonymity)
    • Release a patched version as soon as practical
  • If the vulnerability is declined, we will:

    • Provide a clear explanation of why
    • Suggest alternative resources if applicable

Security Measures

echidnabot implements the following security measures:

  • Webhook Verification: HMAC-SHA256 signature verification for all webhooks
  • Least-Privilege Access: Read-only repository access for cloning, minimal write access for check runs
  • Sandboxed Execution: echidnabot delegates proof verification to ECHIDNA Core; it never executes provers directly
  • Secrets Management: Integration with Vault/SOPS for secure secret handling
  • No Weak Cryptography: SHA-256+ only; no MD5 or SHA-1 for security purposes
  • TLS Everywhere: All external communications use HTTPS/TLS

Security Scanning

This project uses:

  • CodeQL for static analysis
  • TruffleHog for secret detection
  • OpenSSF Scorecard for security metrics
  • ClusterFuzzLite for fuzzing

References

There aren’t any published security advisories