paddle-billing-history

Pass

Audited by Gen Agent Trust Hub on Oct 17, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill demonstrates best practices for data isolation and access control. It explicitly warns developers to retrieve the customer_id from a secure server-side database lookup rather than trusting client-side input, and enforces mandatory filtering in the Paddle API call to prevent unauthorized access to other customers' transaction data. \n- [INDIRECT_PROMPT_INJECTION]: The skill outlines a pattern for ingesting transaction data from an external API (Paddle) for display in a user interface. \n
  • Ingestion points: Data is fetched via paddle.transactions.list in the getBillingHistory Server Action. \n
  • Boundary markers: The skill suggests mapping SDK objects to a simplified Data Transfer Object (DTO) before sending data to the client, effectively limiting the surface area of the exposed data. \n
  • Capability inventory: The logic is restricted to data transformation (formatting currency and dates) and does not involve any shell execution, dynamic code evaluation, or unsafe file operations based on the API response. \n
  • Sanitization: The implementation uses standard numerical parsing and the Intl.NumberFormat API, which treats the ingested data as non-executable strings for display purposes.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 17, 2026, 07:57 PM