Skip to content

Security: aggmoulik/lintscope

Security

SECURITY.md

Security Policy

Reporting a vulnerability

Please do not open a public issue for security problems. Report privately via GitHub's private vulnerability reporting (Security → Report a vulnerability), or email the maintainer listed on the GitHub profile.

We'll acknowledge within a few days and keep you updated through to a fix + disclosure.

Supported versions

lintscope is pre-1.0; security fixes land on the latest published 0.x release.

Threat model (what to look at)

The interesting surface is the local CLI HTTP server that the hosted /studio page talks to. It runs on 127.0.0.1 and is protected by several layers — a bug in any of these is worth reporting:

  • Origin allowlist + CORS — only the configured hosted UI origin may read responses.
  • Per-session token — all data endpoints (/init, /report, /file, /scan, /events) require a crypto.randomUUID() token; /handshake is the only unauthenticated endpoint and is origin/CORS-gated so only the real UI can read it.
  • DNS-rebinding guard — requests with a non-loopback Host header are rejected.
  • Private Network Access preflight handling for public-origin → localhost requests.
  • Path-traversal guardGET /file enforces projectRoot containment (safePath).
  • Data locality — the /studio page must only ever call the local CLI server; it makes no third-party/analytics calls that could see your code or diagnostics.

If you find a way to read a project's source/diagnostics from another origin, exfiltrate the session token, or escape projectRoot, that's a vulnerability — please report it.

There aren't any published security advisories