security-threat-modelот openai

Repository-grounded threat modeling that maps trust boundaries, assets, and abuse paths to concrete code evidence. Enumerates entry points, data flows, and trust boundaries anchored to actual repository structure and configuration Derives realistic attacker goals tied to specific assets (credentials, PII, integrity-critical state, compute resources) rather than generic checklists Prioritizes threats using likelihood and impact reasoning, with explicit assumptions about deployment,...