Skip to content

Security: pavan-dangeti/VeriTrack-AI

SECURITY.md

Security policy

VeriTrack is a personal project that runs only on synthetic data. Security reports are still welcome and taken seriously.

Supported versions

Version Supported
3.0.x yes
< 3.0 no

Reporting a vulnerability

Please report vulnerabilities privately, not in a public issue:

  1. Open the repository's Security tab and choose Report a vulnerability (direct link).
  2. Describe the problem, the affected version or commit, and the steps to reproduce it.

I'll acknowledge the report, keep you informed while it's investigated, and credit you in the advisory unless you'd rather not be named. Please give me a reasonable chance to fix the problem before you disclose it.

Scope

In scope: this repository's code, its CI workflows, and the configuration it ships (Dockerfile, Render blueprints, vercel.json).

Out of scope: the hosted demo's availability (it runs on free plans and sleeps when idle), findings that need an already compromised database or server, and reports from automated scanners without a demonstrated impact.

What the design protects against

docs/threat-model.md describes the assets, the attackers considered, each defence with the test that proves it, and the known gaps (for example, the audit chain cannot detect a fake entry appended at the end). A report about something the threat model already lists as a gap is still welcome if it shows a practical attack.

There aren't any published security advisories