AI-Assisted Secure Coding (Appendix C) · AI Code-Review & Assistant Bot Hardening
Keep attack-testing review bots
Verify that AI review and assistant bots are subject to continuous adversarial testing: indirect-prompt-injection corpora are replayed against the bot through simulated PRs, issues, and comments. Detection effectiveness is tracked over time, and a regression blocks the model or prompt update that caused it.
OWASP AISVS v1.01, requirement AC.11.8
In plain English
Keep testing bots with sets of known injection attacks, replayed through simulated PRs, issues and comments, and block any model or prompt update that makes detection worse.
What good looks like
- Test corpora are replayed regularly.
- Detection rates are tracked over time.
- Regressions block updates.
How teams usually do it
- A scheduled job that opens test PRs in a sandbox repository.
Evidence an auditor would accept
- Detection trends and blocked updates.
Common mistakes
- Testing once at adoption only.
Not sure which requirements apply to your system? Take the assessment.
Requirement text from the OWASP AI Security Verification Standard, licensed under CC BY-SA 4.0. This guidance is shared under the same licence.