fsi-comps-analysis
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external financial tools (MCPs) and user-provided example templates. The absence of defined boundary markers or sanitization logic for this external content creates a vulnerability surface where adversarial data could potentially influence the agent's behavior.
- Ingestion points: The skill instructs the agent to read and reference
examples/comps_example.xlsx, user-supplied templates, and financial data retrieved via FactSet, S&P Kensho, or Daloopa MCP tools. - Boundary markers: The instructions do not define specific delimiters or "ignore embedded instructions" warnings for the external data being processed.
- Capability inventory: The skill utilizes capabilities for automated spreadsheet generation, complex financial modeling, and automated data reporting.
- Sanitization: No evidence of validation, escaping, or filtering of external data before it is incorporated into the analysis output.
Audit Metadata