Skip to content

Security: ferranpons/Llamatik

SECURITY.md

Security Policy

Supported Versions

Llamatik is actively developed, and security updates are provided for recent stable releases.

Version Supported
Latest release (v0.x) ✅ Yes
Previous minor version ✅ Yes
Older versions ❌ No

Only the latest published version (and, when applicable, the immediately previous minor release) receives security fixes. Users are strongly encouraged to upgrade to the latest version.

Reporting a Vulnerability

We take security and privacy issues seriously, especially given Llamatik’s focus on on-device, offline AI.

How to report

If you discover a security vulnerability, please do not open a public GitHub issue.

Instead, report it privately by one of the following methods:

•	GitHub Security Advisories. Use the “Report a vulnerability” button in the repository’s Security tab (preferred).
•	Email (fallback). Send a report to: security@llamatik.com (If email is unavailable, use GitHub Security Advisories.)

What to include

Please include as much of the following as possible:

•	A clear description of the vulnerability
•	Steps to reproduce (proof of concept if available)
•	Affected versions / platforms (Android, iOS, Desktop, library)
•	Potential impact (e.g. crash, data exposure, code execution)
•	Any suggested mitigation or fix (optional)

What to expect

•	Acknowledgement within 72 hours
•	Status updates as the issue is investigated
•	If confirmed, a fix will be prepared and released as soon as reasonably possible
•	Once fixed, we may coordinate responsible disclosure and credit the reporter (if desired)

Scope notes

•	Llamatik runs locally and offline by default; vulnerabilities related to model behavior or prompt content are generally out of scope unless they cause crashes, memory corruption, or unintended data access.
•	Issues in third-party dependencies (e.g. llama.cpp, platform runtimes) may be tracked upstream, but we will document mitigations when applicable.

Thank you for helping keep Llamatik safe and trustworthy 💚

There aren't any published security advisories