shopify-functions
Pass
Audited by Gen Agent Trust Hub on Aug 3, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill defines an indirect prompt injection surface through the ingestion of external data from Shopify Cart and Metafield objects.\n
- Ingestion points: Input data is received via GraphQL queries in
src/input.graphql, which fetch cart details and merchant-defined configuration from metafields.\n - Boundary markers: Code examples for processing metafields use standard parsing methods like
serde_json::from_strorJSON.parsewithout explicit delimiters to isolate configuration payloads from business logic.\n - Capability inventory: Function outputs can modify the checkout and order pipeline state; whitelisted network access is supported via the
allowed_hostsmanifest configuration.\n - Sanitization: Basic type-safe parsing is demonstrated, though the skill does not detail exhaustive validation for untrusted configuration data.\n- [EXTERNAL_DOWNLOADS]: The guide includes instructions for installing standard development tools, such as the
wasm-optoptimizer, using the official Cargo package manager.
Audit Metadata