Case study · Nonprofits

An open development site and an unenforced email policy put donor fraud within reach

Gray-box external red team against the live internet-facing estate, with no application credentials provided. Anonymized.

Engagement
External · Web app · Email
Method
Gray-box · active probing
Validated findings
21
Outcome status
Roadmap delivered; authenticated follow-up test planned
Client profile

Who we worked with.

A nonprofit with a public website, an online course platform and embedded donation pages, run by a small team without a dedicated security function.

Anonymized. No client names, people, domains, addresses or locations are published.

The challenge

The question we were asked.

What can an outsider with no credentials do against the organization's website, its development site, its course platform, its email domain and its donation pages, using live production systems?

What TitanDef did

Assessment modules used.

Authorized external red team, gray-box, against live production systems in a single assessment window.

External red team, gray-box

The live public website and a publicly reachable development copy of it, unauthenticated, with active probing.

Email authentication

The organization's mail domain: sender policy, signing and enforcement.

Course platform and checkout

The online course catalogue, cart and payment flow.

Donation embeds

Third-party donation pages embedded in the website.

Refutation lane

Higher-severity hypotheses actively tested and ruled out before anything was reported.

AI-accelerated, human-validatedAn agent-orchestrated external red team on TitanDef's pipeline, powered by frontier AI models. Every finding was proven live and human-reviewed; five higher-severity hypotheses were tested and refuted, and their evidence is in the report.
What we found

Severity, validated.

21 validated findingsFive attack chains · 5 hypotheses refuted
Critical0
High8
Medium5
Low5
Info3
  • HighA development copy of the website was reachable from the internet with no access gate, exposing staff account names, all media and content, and administrative paths.
  • HighThe domain's email policy did not enforce authentication, so mail spoofing the organization's own address was delivered. Donor-fraud and credential-harvest phishing were viable with zero access.
  • HighCross-origin requests carrying credentials were accepted from any web origin, on both production and the development copy.
  • HighThe course checkout priced every course at zero on the server side, bypassing the payment integration.
  • HighFour more High findings: unauthenticated media and content exposure on the development copy, and three outdated website components with published vulnerabilities, held for confirmation in an authenticated follow-up test.
Outcome

What we recommended.

The most significant observation: an unenforced email policy plus an unprotected development site together gave an immediately exploitable path to donor fraud and full staff-account enumeration, without compromising any production system. Two low-effort changes collapse most of it: enforcing the email policy removes the zero-credential phishing chain, and gating the development site removes the unauthenticated surface behind two more chains.

0–30 days

Enforce the email policy in quarantine mode; gate the development site; restrict cross-origin access to an allowlist; update the three outdated components.

30–90 days

Move the email policy to reject after reviewing reports; confirm course pricing and validate the payment integration; add login rate limiting.

Follow-up

An authenticated test to confirm the three deferred component findings.

Deliverables

What the client received.

  • 21 validated findings with evidence and remediation steps
  • Five scored attack chains, ranked by exploitability today
  • Refuted hypotheses: what was tested and ruled out
  • Detection and response guidance for the organization's IT team
  • Prioritized remediation roadmap with effort estimates

Prepared by TitanDef offensive security team

Engage

Find out what an adversary already sees.

Focused engagements run 5 to 7 business days, with a 72-hour first-findings commitment. Powered by frontier AI models; every finding validated by a senior human.