Security context
What an agent needs to avoid regressing past fixes and find the next vuln in this repo.
intuit/karate
main @ 2534455
3
Fixes
0
CVEs
HIGH
Peak severity
Highlights
Information Disclosure: 1 prior fix. Scrutinize any change in this area.
io.karatelabs.core.MockHandler: most-fixed (1 issue). Treat as high-risk during review.
1 high-severity fix in this history; regressions here are high-impact.
Recurring patterns
The bug types that recur here, drawn from past fixes, not open vulnerabilities.
Remote Code Execution: The mocking engine's evaluation of request data allowed attackers to achieve arbitrary code execution via Java interop. Developers must ensure that Java interop and dynamic expression evaluation on untrusted request payloads remain completely disabled by default in mock handlers.
Information Disclosure: Sensitive HTTP headers like Authorization tokens can leak in plaintext to console and report logs during nested feature executions (e.g., callonce in Background steps) if the caller thread's LogContext is not explicitly restored.
Reflected XSS: Untrusted input or extension-controlled page titles rendered directly into HTML reports can execute malicious scripts if not HTML-encoded. Ensure all output parameters in report generators are strictly sanitized before rendering.