Skip to content

Security: armanas/froglet

Security

SECURITY.md

Security Policy

Reporting a Vulnerability

If you discover a security vulnerability in Froglet, please report it responsibly by emailing security@armanas.dev.

Please include:

  • A description of the vulnerability and its impact
  • Steps to reproduce or a proof-of-concept
  • The affected version(s)

What to expect

  • Acknowledgement within 48 hours of your report.
  • An assessment within 7 days: severity, affected surfaces, and whether we agree it is a vulnerability.
  • A fix or documented mitigation within 90 days for confirmed issues — usually much sooner; this is a small codebase with a single maintainer and security fixes jump the queue.
  • Coordinated disclosure: we ask that you not publish details until a fix has shipped or 90 days have elapsed, whichever comes first. We will credit you in the changelog and release notes unless you prefer otherwise.
  • If a report affects the hosted services (ai.froglet.dev, marketplace.froglet.dev, try.froglet.dev, arbiter.froglet.dev), we may deploy a mitigation there before the open-source fix lands; the public fix still follows the timeline above.

Scope

This policy covers:

  • The Froglet protocol implementation and kernel (froglet-protocol, conformance/)
  • The reference node (froglet-node, provider and runtime planes, settlement drivers including Lightning, Stripe, and the experimental x402 driver)
  • The publish engine and bootstrap/install scripts
  • The OpenClaw/NemoClaw and MCP integrations and the shared JS client library
  • The hosted first-party deployments listed above

Out of scope: vulnerabilities requiring physical access, social engineering of the maintainer, volumetric denial-of-service against the hosted free tier, and issues in third-party dependencies that are already public (report those upstream; we track them via cargo audit / npm audit in CI).

Threat model

The protocol's trust boundaries, key-compromise blast radius and response runbook, and known accepted risks are documented in docs/THREAT_MODEL.md. Read it before reporting — it also tells you which behaviors are deliberate (e.g. non-refundable base fees, operator-adjudicated marketplace enforcement).

Supported Versions

Only the latest release on the main branch is actively supported with security fixes. Signed froglet/v1 artifacts remain verifiable across versions per docs/VERSIONING.md, so upgrading the node never invalidates existing evidence.

There aren't any published security advisories