Perform CWE-grounded security code reviews and precise weakness mapping using a locally derived MITRE CWE corpus, relationship graphs, mapping notes, detection methods, mitigations, and schema semantics. Use when Codex needs to audit source code or pull requests, identify root-cause weakness classes, distinguish broad symptoms from mappable CWEs, justify CWE IDs in findings, or review code against CWE views such as Software Development, Research Concepts, Top 25, OWASP, language-specific, or AI/ML weakness sets.