payout-split-management
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the processing of untrusted external data (orders, seller metadata) through a pipeline that includes sensitive financial and database operations.\n
- Ingestion points: Javascript snippets in
SKILL.mdingest and process parameters from external orders, includingorderTotalCents,sellerEmail,sellerId, andorderSubtotal.\n - Boundary markers: The code samples and instructions do not include boundary markers or explicit guidance for the agent to disregard instructions potentially embedded in metadata or order fields.\n
- Capability inventory: The skill utilizes capabilities with high impact, specifically financial disbursements via the Stripe API (
stripe.transfers.create) and modification of seller records in the database (db.sellers.update).\n - Sanitization: No sanitization, validation, or escaping logic is demonstrated for untrusted inputs before they are passed to sensitive API or database calls.
Audit Metadata