shipping-cost-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 external data sources such as CSV exports and connected store analytics (MCP), which creates a surface for potential indirect prompt injection.
  • Ingestion points: Order-level exports, carrier cost figures, and zone data are ingested as specified in the 'How to run' section of SKILL.md.
  • Boundary markers: The skill requires the agent to run data integrity checks from references/data-input-integrity.md before performing any calculations.
  • Capability inventory: The skill's capabilities are restricted to data processing, market benchmarks via WebSearch, and rendering visual bar charts; it does not possess capabilities for arbitrary command execution or network-based data exfiltration.
  • Sanitization: The methodology explicitly instructs the agent to identify and exclude unreadable or negative order values to ensure the reliability of the analysis.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 25, 2026, 04:06 PM