RAK OS is infrastructure for media without an owner. Its integrity — provenance, the reader tier, node isolation — is part of the mission, so we take security reports seriously and credit the researchers who help (the Warden role in REALM.md).
Do not open a public issue for a security vulnerability.
Report privately via GitHub's private vulnerability reporting (Security → Report a vulnerability), or by direct message to the Crown (@Hei33enberg).
Please include: what you found, how to reproduce it, and the potential impact. We aim to acknowledge within a few days.
In scope:
- The
@rak-ad/mcpclient and this repository. - The RAK content object / provenance model as specified in
SPEC.md(e.g. signature or tamper-evidence weaknesses). - Tier/scope enforcement described in the skill docs (e.g. a reader-tier caller reaching a paid or internal tool).
Out of scope (report to rak.guru directly, not here): the hosted rak.guru product surface, third-party dependencies with their own advisories, and volumetric/DoS testing against the live node.
- We will not pursue legal action against good-faith research that respects this policy and user privacy.
- We disclose fixed vulnerabilities and credit the reporter (unless anonymity is requested).
- We make no absolute security claims. The honest posture of what is and isn't protected lives in the code and the spec — verify it there.