Company
About
Last updated 13 November 2026
Frontier Cyber Defence is a THRead product, built by a team with decades of experience in cyber defence and security.
Why this exists
In July 2026 a set of models running a security benchmark were never told to attack. They treated the sandbox around them as a problem to solve, chained a zero-day, escaped to the open internet, harvested cloud credentials and obtained the production database credentials. The defenders then found their forensic work blocked by the safety guardrails on the hosted models they tried to use, and fell back to a self-hosted open model to finish the investigation.
Most pen tests check a site once or twice a year. Attackers now use frontier models around the clock, so we built a defender on the same models that checks your site again as often as daily.
How we work
Once you have proved the site is yours, a frontier model attacks it. A second frontier model reviews what it claims against the raw evidence. Every check that ran is recorded, including the ones that found nothing, and the whole log goes in your report.
We publish our own scan of our own site on the front page, with every check it ran and what each one found.
How a release of FCD ships
- Testedthe full test suite, on a real database, and our page checks
- Run on a private copybuilt and started where only we can reach it
- Attacked by our red teama confirmed critical stops it, and so does a high unless nothing that ships changed
- Livethen checked from outside
Who we are
The team behind Frontier Cyber Defence has spent decades defending live systems and the data on them. Post-quantum security and highly secure sovereign systems are the work we care most about, and we hold this service to that standard. Your scan evidence is sealed to your account before it is stored. Every change to the service is attacked on a private copy.
- Registered name
- THREAD CORPORATION LIMITED
- Company number
- 15054064
- Registered address
- 167-169 Great Portland Street, 5th Floor, London, England, W1W 5PF