Skip to content

Security: defenseunicorns/peat

SECURITY.md

Security Policy

Reporting a Vulnerability

If you discover a security vulnerability in peat, please report it responsibly. Do not open a public GitHub issue for security vulnerabilities.

How to Report

You have two options:

  1. Email: Send a detailed report to security@defenseunicorns.com
  2. GitHub Security Advisories: Use the private vulnerability reporting feature on this repository

What to Include

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

Response Timeline

  • Acknowledgment: Within 3 business days
  • Initial assessment: Within 10 business days
  • Fix timeline: Dependent on severity

Disclosure Policy

  • We will acknowledge reporters in the remediation PR (unless anonymity is requested)
  • We follow coordinated disclosure practices
  • We aim to release patches before public disclosure

Supported Versions

Version Supported
latest Yes

Security-Relevant Areas

peat is a tactical mesh protocol workspace comprising 10 crates. The following areas are particularly security-sensitive:

  • Protocol security: Formation keys, membership certificates, and channel encryption govern mesh access control. Weaknesses here could allow unauthorized nodes to join a formation or eavesdrop on traffic.
  • CRDT sync integrity: Automerge and Ditto backends replicate state across peers. Malicious or malformed CRDT operations could corrupt shared state or cause divergence.
  • Transport security: QUIC/Iroh, BLE, and UDP bypass channels carry mesh traffic. TLS configuration, connection establishment, and stream multiplexing must enforce secure defaults across all transports.
  • FFI boundary: peat-ffi exposes mobile bindings. Memory safety, input validation, and error handling at the FFI layer are critical to prevent undefined behavior in host applications.
  • TAK bridge: CoT protocol translation bridges tactical data between peat and TAK ecosystems. Malformed or spoofed CoT messages could inject false situational awareness data.
  • Edge inference: Model distribution and orchestration move ML artifacts across the mesh. Tampered models or unauthorized orchestration commands could compromise edge nodes.

Security Best Practices

When integrating peat, follow these practices:

  • Use certificate-based enrollment for production deployments
  • Enable formation key authentication to restrict mesh membership
  • Use unique formation secrets per mesh network to prevent cross-network access
  • Deploy behind TLS-terminating ingress for REST APIs
  • Keep dependencies up to date

There aren’t any published security advisories