We Built an AI Agent That Intentionally Breaks Code to Fix It Before Hackers Do

Team Syntax and Synergy built Resilix, an autonomous adversarial agent that intentionally breaks code to surface vulnerabilities and auto-generates defensive security patches before code reaches production, as their submission to the IBM Bob 2.0 Hackathon. The tool targets unhandled null pointers, string overflows, and SQL injections that slip through under hostile edge cases, aiming to replace slow late-stage pipeline audits with real-time defensive assistance. The entire application was developed inside the IBM Bob 2.0 environment, including a fuzzing API endpoint, scoring formulas, a single-page interactive UI, and tracked execution logs.

We just wrapped up our submission for the IBM Bob 2.0 Hackathon Our team, Syntax and Synergy , built Resilix —an autonomous adversarial agent that tests code resilience and auto-generates defensive security patches before code hits production. Developers and AI models are writing code faster than ever. However, security testing tools are often built for slow, late-stage pipeline audits rather than active development. Unhandled null pointers, string overflows, and SQL injections frequently slip through under hostile edge cases. Resilix acts as a real-time defensive assistant: The entire application was built inside the IBM Bob 2.0 environment: /api/fuzz , and scoring formulas. app.py and single-page interactive UI index.html . .bob/rules- and tracked all execution logs in bob sessions/ . We'd love to hear your thoughts and feedback from the DEV community 🚀