HumanAIOS is an integrity-measurement project. Holding ourselves to the security standard we ask others to meet is part of the Witness principle: the scrutiny we apply to AI substrates applies equally to the systems and operators running the assessments.

Please report privately — do not open a public issue for a security matter.

1. Preferred: use GitHub's private vulnerability reporting on this repo — Security → Advisories → “Report a vulnerability.” This opens a private channel with the maintainers.
2. If that is unavailable, contact the maintainer through their GitHub profile (@humanaios-ui) and request a private channel before sharing details.

Please include: what you found, where (repo + path/URL), how to reproduce, and the impact you see. We aim to acknowledge within 72 hours and to agree a remediation and coordinated-disclosure timeline with you from there.

This is a pre-launch R&D ecosystem; the most sensitive surfaces are:

• The ACAT pipeline (acat/) and any assessment write path — corpus integrity is a research-validity concern as well as a security one. Report anything that lets unauthenticated writes reach acat_assessments_v1 or the derived stats.
• The site generator (site/, tools/registry_site_generator*) — the public site must never publish content marked internal-only. A rendering path that leaks private registry entries or PII is in scope.
• The document-control system (.doc-control/, document-registry.yaml) — a bypass that lets a controlled document be altered without tripping the gate.
• Any workflow / CI misconfiguration (see .github/workflows/).

Out of scope: findings that require operator-level (Zone 3) credentials you were given legitimately; volumetric/DoS testing against live services; social engineering.

No secret is ever committed to a repository. Credentials live in a secrets manager or hardware key — never in files, a browser, or an AI context, and we use short-lived cut credentials rather than minting long-lived ones. Handling of credential-bearing actions is Zone 3 (operator-only) under GOVERNANCE.md.

If you find a committed secret (even a placeholder that invites a real one), report it privately as above. We run toward adopting automated secret scanning (GitHub secret scanning + gitleaks) as part of the health growth path in SYSTEM_HEALTH.md.

main is the living line and the only supported branch. Fixes land on main; there are no maintained release branches at this stage.

_{◐ The Witness · security scrutiny applies in both directions. See SYSTEM_HEALTH.md for the wider health posture.}