mortgage-loan-fallout-premortem
Pass
Audited by Gen Agent Trust Hub on Aug 22, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill is designed to process and analyze untrusted external data such as borrower income statements, credit reports, and asset documentation. This creates a surface for indirect prompt injection where instructions embedded in processed documents could attempt to subvert the agent's logic. * Ingestion points: Mortgage file submissions, underwriting documents, and borrower financial records (SKILL.md). * Boundary markers: None identified in the skill instructions. * Capability inventory: None. The skill is instructional and does not provide accompanying scripts or tool definitions. * Sanitization: None identified; the skill assumes direct analysis of provided data.
- [EXTERNAL_DOWNLOADS]: The skill includes informational links to the author's official website (deciqai.com) and GitHub organization (github.com/deciqai) for extended documentation, examples, and machine-readable metadata. These references target vendor-owned infrastructure.
Audit Metadata