Vulnerability Disclosure Policy
Last updated: 1 September 2026
SARGOS is an incident-coordination platform used during search-and-rescue operations. We take the security and privacy of responder and subject data seriously, and we welcome reports from security researchers acting in good faith. This policy explains what is in scope, how to report, and what you can expect from us.
Scope
In scope:
getsargos.com(andwww.getsargos.com), the web application.api.getsargos.com, the backend-for-frontend (BFF) API.
Out of scope (please do not test these):
- Denial of service (DoS/DDoS), volumetric, or resource-exhaustion testing: this service is used during live rescue operations; degrading it can disrupt an active response.
- Social engineering of staff, contractors, customers, or responders (phishing, vishing, pretexting), and any physical attack against people, offices, or hardware.
- Attacks requiring a compromised device or account, or physical access.
- Findings from automated scanners with no demonstrated, exploitable impact (please triage before reporting).
- Third-party services we depend on (Supabase, Fly.io, Vercel, Cloudflare, Stripe, Resend, Mapbox, Sentry, Inngest): report those to the respective vendor. If our configuration of one of them is the flaw, that is in scope.
- Missing security headers and best-practice suggestions with no concrete exploit, spam or content issues, and reports based solely on outdated-software version banners.
If you are unsure whether something is in scope, ask first. See How to report.
Safe harbor
If you make a good-faith effort to comply with this policy during your research, we will consider your research authorized, we will not pursue or support legal action against you for it, and we will work with you to understand and resolve the issue quickly. “Good faith” means you:
- Follow this policy and stay within the in-scope targets.
- Avoid privacy violations and harm. Do not access, modify, exfiltrate, or store data that is not your own. A single record is enough to demonstrate most issues. Stop immediately if you encounter personal data of responders or rescue subjects, and tell us.
- Do not degrade, disrupt, or deny service, and do not run intrusive automated scanning.
- Give us a reasonable time to remediate before any public disclosure (see below).
What this authorization covers
For research that stays inside this policy, we authorize you in advance to:
- Access our systems in ways that would otherwise be “without authorization” or “exceeding authorized access” under the U.S. Computer Fraud and Abuse Act (18 U.S.C. §1030) and the equivalent state and foreign computer-misuse laws.
- Circumvent a technological access control on our own code or data where doing so is necessary to find, confirm or demonstrate a vulnerability. We will not bring, or support, a claim under the anti-circumvention provisions of the DMCA (17 U.S.C. §1201) or their foreign equivalents for that conduct.
- Do the above notwithstanding our Terms of Service and Acceptable Use Policy. Good-faith testing under this policy would otherwise breach parts of those documents: the Acceptable Use Policy prohibits probing security controls and circumventing access controls, and it means it for everyone who is not doing this. For the scope and duration of authorized testing, this policy controls and we waive that breach.
That waiver does not reach the availability restrictions. Denial-of-service, volumetric and resource-exhaustion testing stay out of scope and stay a breach, for the reason given above: this service is used during live rescue operations.
What this safe harbor is not
It is an undertaking about our own claims, and nothing wider:
- It binds only us. It does not bind our customers, the organizations whose data we host, our vendors (Supabase, Fly.io, Vercel, Cloudflare, Stripe, Resend, Mapbox, Sentry, Inngest), or any law-enforcement or government body. None of them is a party to this policy.
- It is not an indemnification. We do not agree to defend you, to pay your legal costs, or to cover any claim, fine, or damages a third party brings against you arising from your research. What we will do, if a third party acts against you for activity conducted under this policy, is make this authorization known to them, and to a court if you ask us to. That is the whole of the commitment, and we would rather state its limit plainly than have you discover it later.
- It covers good faith, including a good-faith mistake. If you step outside this policy accidentally, stop, and tell us promptly, we will treat the research as authorized. It does not cover data theft, extortion, service disruption, or continued testing after we have asked you to stop.
How to report
- Email: security@getsargos.com
- Please include: a clear description, the affected URL or endpoint, reproduction steps (a proof-of-concept if you have one), and the impact you believe it has. Screenshots or a short screen recording help.
- Encryption: if you need to send sensitive details, ask in your first message and we will arrange a secure channel. We publish no PGP key today, and
security.txtcarries noSignature:field for the same reason: a key we do not hold is worse than none. - Please report in English.
Do not include live exploitation of real user data in your report; a redacted proof-of-concept against a single record is sufficient.
What to expect from us
We are a small team, so these timelines are honest rather than aspirational:
| Stage | Target |
|---|---|
| Acknowledge your report | within 3 business days |
| Triage + initial severity assessment | within 10 business days |
| Status updates while we work | at least every 2 weeks until resolved |
| Remediation | prioritized by severity; field-critical and data issues first |
We will let you know when the issue is resolved, and (if you would like) we are happy to credit you.
Coordinated disclosure
- Please keep the vulnerability confidential until we have remediated it and agreed on a disclosure timeline.
- Our default coordinated-disclosure window is 90 days, counted from the day we validate the report (the day we confirm the issue is real and in scope), not from the day you sent it and not from the day we acknowledged it. We will tell you that date when we set it. Acknowledgement is a receipt; validation is the first day on which a remediation clock can honestly run, and anchoring the window there means a slow triage on our side eats our own remediation time instead of yours. It may of course be sooner by mutual agreement once a fix is deployed.
- Active exploitation shortens this window; it never lengthens it. If either of us finds evidence that the vulnerability is being exploited in the wild, the 90-day default stops applying and we move to whatever timeline protects responders and rescue subjects, which may mean issuing an advisory or notifying affected organizations before a fix is complete. We will coordinate with you where time allows, and tell you either way.
- If remediation is taking longer than the window, we will tell you why and work out a revised timeline with you. We will not use the window to sit on a fix.
- Please give us the chance to confirm the fix before you publish, and avoid disclosing details that could expose responder or subject data even after remediation.
Recognition
We do not currently run a paid bug-bounty program. We are grateful for responsible reports and will gladly acknowledge researchers who report valid, in-scope issues in good faith (with your permission). We keep no public acknowledgements page yet, so security.txt carries no Acknowledgments: field; ask in your report and we will credit you wherever the fix is announced.
The machine-readable version of this contact record is served at /.well-known/security.txt (RFC 9116).