saas-replacement-planner
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill workflow involves parsing untrusted external inputs such as CSVs and bank statements, which can contain adversarial instructions. \n
- Ingestion points: The skill processes external CSVs, screenshots, and bank statements as defined in
SKILL.md. \n - Boundary markers: No boundary markers or instructions to ignore data-embedded commands are present. \n
- Capability inventory: The skill has access to powerful tools including
Bash,WebSearch,WebFetch,Write, andEdit. \n - Sanitization: There is no evidence of input validation or sanitization of data extracted from external documents.
Audit Metadata