Report privately through GitHub. Please do not open a public issue for a security problem.

1. Open the Security tab of this repository
2. Click Report a vulnerability

That creates a private advisory visible only to the maintainers. Private vulnerability reporting is enabled on this repository.

If you cannot use GitHub, open a public issue saying only that you have a security report and how to reach you — no details.

• What an attacker can do, and what access they need in order to do it
• The file and line, or a URL and the steps to reproduce
• Whether you are reporting against the published site or a local checkout

• Acknowledgement within 5 working days
• An assessment, and either a fix or an explicit decision not to fix, within 30 days
• Credit in the advisory, if you want it

There is no bug bounty. The Ethical Tech CoLab is a small research group and this is unfunded work; please be patient.

There are no releases or version tags. Only the current state of the main branch is supported, and fixes are not backported.

ERCF — the Evacuation Risk and Cost Framework: a seven-dimension severity model with a cost engine and a FastAPI service, plus a browser dashboard.

In scope

• Authentication and authorisation flaws — anything that lets a caller act as another user, or reach an endpoint they should not
• Injection of any kind, and cross-site scripting in rendered output
• Any path by which a server-side credential becomes reachable from a client
• Server-side request forgery, or a request whose destination a caller controls
• Uncapped or unauthenticated access to a paid third-party API — cost is a real impact here, not a theoretical one
• Exposure of a credential or token in committed code

Out of scope

• Findings that require access to a maintainer's machine, or to the deployment's environment variables
• Rate-limit tuning, absent a concrete abuse path
• Denial of service by request volume alone

Twelve CDN <script> tags are currently loaded without a Subresource Integrity attribute. A fix is pending in PR #3.

Disagreement with the model, the numbers, the methodology, or the legal analysis is not a security report — but we do want to hear it. Open a normal issue. If a peer review is published in this repository (PEER-REVIEW.md), read it first: the finding may already be recorded there.

These are research outputs and prototypes, published to show method and to be argued with. They are not production systems and should not be used to make real operational decisions.