Skip to content

Security: sudoeren/localce

Security

.github/SECURITY.md

Security Policy

Supported Versions

The following versions of Localce receive security updates:

Version Supported
main
latest
< 1.0

Reporting a Vulnerability

Localce is a privacy-first project. If you discover a security vulnerability, please report it privately so we can address it before public disclosure.

Please do NOT open a public GitHub issue for security vulnerabilities.

How to report

Use one of the following channels (in order of preference):

  1. GitHub Security Advisories (recommended) Go to the Security tab of this repository → "Report a vulnerability" → fill in the form.

  2. Email Send details to the maintainer listed in CODEOWNERS or the repository owner profile.

What to include

  • A clear description of the vulnerability and its impact
  • Steps to reproduce (proof-of-concept preferred)
  • Affected version/commit
  • Any known mitigations
  • Your name/handle for credit (optional)

What to expect

  • Initial response: within 72 hours
  • Status update: within 7 days
  • Patch timeline: critical issues are prioritized; we aim to ship a fix within 30 days

We follow coordinated disclosure and will credit reporters in the release notes (unless you prefer to remain anonymous).

Security Design Notes

Localce is designed with the following security principles:

  • Local-only execution — All translations happen on your machine via LM Studio. No text is ever sent to remote servers owned by Localce.
  • No telemetry — Next.js telemetry is explicitly disabled in the Docker image (NEXT_TELEMETRY_DISABLED=1).
  • Non-root container — The Docker image runs as a dedicated unprivileged nextjs user.
  • Outbound network — The only outbound request the app makes is to the user-configured LM_STUDIO_URL (default: http://localhost:1234).

If you find that the application unexpectedly opens network connections to a destination other than LM_STUDIO_URL, please report it.

Scope

In scope:

  • Cross-site scripting (XSS) in the web UI
  • Server-side request forgery (SSRF) in the translation API route
  • Arbitrary code execution via crafted model responses
  • Container escape / privilege escalation
  • Dependency vulnerabilities with realistic exploit paths

Out of scope:

  • Issues in third-party models running in LM Studio
  • Issues requiring physical access to the user's machine
  • Denial of service against the user's own machine

Thank you for helping keep Localce and its users safe.

There aren't any published security advisories