Skip to content

Security: TursiThePanda/NDU

Security

SECURITY.md

Security Policy

Supported versions

NDU is a hobby project with a single maintained line. Security fixes are applied to the latest tagged release on main. Older tags receive no backports.

Reporting a vulnerability

Please report suspected vulnerabilities privately, not in a public issue. Two channels:

  1. GitHub private advisory — preferred. Open an advisory directly on the repo. Only maintainers see it until publication.
  2. Email — if GitHub advisories are unavailable to you, reach out via the maintainer's contact info on the GitHub profile.

Please include:

  • a description of the issue and its impact,
  • reproduction steps or a proof-of-concept,
  • affected versions (commit SHA or tag),
  • any suggested mitigation.

Expect an initial acknowledgment within 7 days. Fix timelines depend on severity and the maintainer's availability — this is a spare-time project.

Scope

In scope:

  • Privilege escalation via NDU's code paths (auth bypass, session handling, admin-only endpoint exposure).
  • Unauthenticated or unintended data exposure (device metadata, credentials, session tokens).
  • Injection flaws (command, SQL, XSS in the admin UI).
  • Credential storage issues (data/users.json, custom/config.json, providers/*/config.json).
  • Path traversal in any endpoint that accepts filenames.

Out of scope:

  • Issues that require root on the host NDU runs on (by design, the service user has access to its own configs and data).
  • Issues in third-party dependencies with an already-disclosed CVE — please report those upstream.
  • Denial-of-service via unauthenticated burst traffic to /api/* (not addressed today; expected behaviour on a LAN service without a rate-limit layer in front).
  • Social-engineering or physical-access attacks.

Hardening notes

NDU ships scripts/install.sh which sets file permissions (chmod 600 on config files containing secrets, chmod 700 on data/) and optionally creates a dedicated ndu service user. See docs/secure-deployment.md for the threat model and step-by-step hardening.

There aren't any published security advisories