supply-chain-optimization-shopify
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied business metrics (e.g., product costs, shipping fees) to generate Markdown and HTML reports. This creates a surface for indirect prompt injection if malicious instructions were embedded in input fields, though the risk is minimal as the tool primarily processes numeric data.
- Ingestion points: Business data is ingested through the
SupplyChainInputdataclass inscripts/calculator.py, which is populated via CLI arguments. - Boundary markers: None present; the skill lacks explicit markers to distinguish data from instructions in generated reports.
- Capability inventory: The skill has file-writing capabilities through
scripts/report_doc.pyandscripts/report_html.py, which generate the filesreport.mdandreport.htmlrespectively. - Sanitization: The skill relies on numeric type enforcement in the
SupplyChainInputdataclass, but does not sanitize string fields used in report generation. - [SAFE]: The skill fetches the
chart.jsvisualization library from jsDelivr, a well-known and trusted CDN service. This is standard practice for including legitimate client-side dependencies for data visualization in the HTML report.
Audit Metadata