Skip to content

Security: Chevron7Locked/lm-chat

Security

SECURITY.md

Security Policy

Reporting a Vulnerability

Email security@chevron7.io with:

  • Description: What is the vulnerability?
  • Affected version: Which release(s)?
  • Proof of concept: Reproduction steps or a test case
  • Severity: Your assessment (critical / high / medium / low)

Response timeline:

  • Acknowledgement within 48 hours
  • Status update within 7 days
  • Fix and coordinated disclosure once patched
  • Credit in the release notes (unless you prefer anonymity)

Please do not publicly disclose before a patch is available.

Scope

LM Chat is self-hosted and single-instance by design — a handful of trusted people share one deployment. In scope:

  • Authentication and session handling, TOTP two-factor, the invite flow
  • Per-user data isolation (chats, projects, documents, memory)
  • Provider-credential handling (encrypted at rest, never returned to the browser)
  • The MCP tool runtime and its filesystem sandbox
  • SSRF, injection, and the streaming/upload surfaces

Out of scope: issues that require an already-compromised admin account or host, and anything in the internal docs/ tree (not shipped).

Supported versions

Security fixes land on the latest release. Pin to a released tag (ghcr.io/chevron7locked/lm-chat:<version>) rather than :latest if you need reproducible deployments.

There aren't any published security advisories