api-graphql
Fail
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: HIGHPROMPT_INJECTIONNO_CODE
Full Analysis
- [PROMPT_INJECTION] (HIGH): The skill establishes an ingestion surface for untrusted external data via Shopify GraphQL queries.
- Ingestion points: GraphQL queries in
references/QUERY_EXAMPLES.md(e.g.,GetOrders,GetCustomers) retrieve sensitive PII into the agent's reasoning context. - Boundary markers: No delimiters or safety instructions are present to prevent the agent from interpreting ingested data as commands.
- Capability inventory: The mutation templates for
productUpdate,inventoryAdjustQuantities, andwebhookSubscriptionCreateprovide high-privilege write and exfiltration capabilities. - Sanitization: No validation, escaping, or filtering is defined for the variables or the returned external data.
- [NO_CODE] (INFO): No executable code, shell scripts, or binary files were found; the file consists of documentation and data templates.
Recommendations
- AI detected serious security threats
Audit Metadata