AIMoat
I built a browser-based scanner for public AI chatbot integrations. It combines configuration checks and interaction probes with severity scoring, evidence and remediation-oriented reporting.
The system behind the interface.
AI security claims are easy to overstate. A useful scanner has to distinguish public configuration evidence from platform vulnerabilities and present reproducible findings.
Architecture choices that shaped the product.
The scanner focuses on what a browser and an unauthenticated visitor can observe instead of claiming access to private model or platform internals.
A grade is never the only result. Findings retain the request, response, browser signal, or configuration detail that produced them.
Headers and exposed data need deterministic inspection; model-behaviour risks need carefully scoped interaction probes.
What I developed.
- Static checks across scripts, browser storage, headers, and exposed endpoints
- Dynamic prompt-injection and prompt-extraction probe corpus
- Severity scoring with masked and full-evidence report modes
The resulting system.
A working scanner engine tested against public chatbot integrations, with platform detection, multiple check modules, attack probes, and report generation.
Relevant to your project.
Browser automation, AI integration testing and reporting tools.