Skill UI

This skill focuses on identifying security vulnerabilities in Large Language Model (LLM) and AI features. It covers prompt injection, system prompt leakage, cross-tenant access, and data exfiltration techniques. Users must verify findings against false positives using reproducibility checks and out-of-band callbacks before reporting.

Finds similar vulnerabilities and bugs across codebases using pattern-based analysis. Ideal for hunting bug variants, building CodeQL/Semgrep queries, analyzing security issues, and systematic code audits after an initial discovery.