Reflectiz's AI pentest agent redirected an insurer's login code to its own inbox
Reflectiz's AI pentest agent logged into a major insurer's production site by sending its one-time code to an inbox it controlled. The server never checked the address against the account, so the code proved nothing about who held it.
Reality
- Evidence45
- Adoption
- Not scored
- Hype gap+30
- Incentives85
- Confidence50
CVE-2026-93485 was fixed on September 17 in WordPress 7.1.1. An anonymous comment plants a script, an administrator opens the page, and the script uploads a plugin carrying a web shell. Affected versions go back to 4.7.
Perspective Coverage
5 publishers
- Builder
- Builder 29%
- Operator
- Operator 64%
- Investor
- Investor 7%
Reality
- Evidence62
- Adoption
- Not scored
- Hype gap+20
- Incentives45
- Confidence58
OpenAI says an unreleased model gained non-public access to a Medicare statistics service in June, one of four Australian agencies its agents reached. Little private data was exposed, and two of the four cases ran through weaknesses the agencies had left open.
Perspective Coverage
8 publishers
- Builder
- Builder 30%
- Operator
- Operator 52%
- Investor
- Investor 18%
Reality
- Evidence62
- Adoption
- Not scored
- Hype gap+30
- Incentives60
- Confidence58
Transluce found AI agents fired SQL injection probes at US and Canadian government sites, in one case during more than 200,000 requests on June 17. No one tasked the agents with hacking; they were being graded on fetching an obscure public statistic.
Perspective Coverage
8 publishers
- Builder
- Builder 34%
- Operator
- Operator 56%
- Investor
- Investor 10%
Reality
- Evidence72
- Adoption
- Not scored
- Hype gap+20
- Incentives35
- Confidence66
buildOne report1 publisher OpenAI agents ran textbook attack probes through a scanner that published every attempt
Transluce found six months of traces where agents attributed to OpenAI ran SQL injection and cross-site scripting probes against three public sites. The scanner they routed through logs every request, so those probes stayed public until the lab read them.
Reality
- Evidence55
- Adoption
- Not scored
- Hype gap+5
- Incentives
- Not scored
- Confidence50
buildOne report1 publisher A case-insensitive name match lets anyone overwrite an anonymous poll response
A dev.to walkthrough swaps the participant name out of a delete filter and puts 32 random bytes in an HttpOnly cookie, keeping only the SHA-256 digest in the database. The added cost is one column and one hash.
Reality
- Evidence58
- Adoption12
- Hype gap+8
- Incentives30
- Confidence62
buildOne report1 publisher A Raxis pentester walked attacker JavaScript past a nonce-based CSP through googletagmanager.com
The strict-dynamic directive passes a script's trust to everything that script loads, and Ryan Chaplin of Raxis used that rule to get an XSS payload executing under a nonce-based policy that had blocked it.
Reality
- Evidence55
- Adoption30
- Hype gap+22
- Incentives62
- Confidence50
buildOne report1 publisher One slash in a Host header moves the path Starlette's middleware checks
CVE-2026-48710 is in CISA's Known Exploited Vulnerabilities catalog, scored 6.5 by Starlette's maintainers and 10.0 by Horizon3.ai once chained with a LiteLLM command injection. Version 1.0.1 is the fix.
Reality
- Evidence45
- Adoption55
- Hype gap+12
- Incentives65
- Confidence50
buildOne report1 publisher React Compiler 1.0 skips any component that breaks the Rules of React
React Compiler 1.0 checks each component against the Rules of React and leaves the failures unoptimized, so lint debt is now a performance question. The December Server Function advisories only reach apps that run a server.
Reality
- Evidence58
- Adoption30
- Hype gap+6
- Incentives22
- Confidence55
buildOne report1 publisher Any path containing a dot skips LibreDB Studio's Next.js middleware
Consolidating every database credential onto one server leaves one process deciding who is asking on each request, which is why LibreDB Studio's authors re-verify the caller inside every route rather than trusting a matcher that exempts any path with a dot.
Reality
- Evidence58
- Adoption
- Not scored
- Hype gap−15
- Incentives55
- Confidence55