skills/spree/agent-skills/spree-taxes/Gen Agent Trust Hub

spree-taxes

Pass

Audited by Gen Agent Trust Hub on Oct 1, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill describes an architecture for processing external data that presents a potential surface for indirect prompt injection.
  • Ingestion points: The skill instructs on handling tax_identifier (VAT numbers), exemptions, and customer metadata via the set_tax_line_context hook and provider estimate method in SKILL.md.
  • Boundary markers: The documentation does not specify the use of delimiters or specific instructions for the agent to disregard potential instructions embedded within these data fields.
  • Capability inventory: Custom tax providers documented in SKILL.md have capabilities to perform database writes (Spree::TaxLine.create!), execute external network calls (MyTax::Client.quote), and impact the order finalization workflow.
  • Sanitization: While the skill mentions format-based validation for VAT identifiers, it does not detail sanitization or escaping of the input values before they are processed by the provider logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 1, 2026, 02:27 PM