customer-management
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines patterns for retrieving customer records and wallet ledger entries from the Dodo Payments API, which serves as an ingestion point for untrusted data provided by external users (e.g., metadata, names, or transaction reasons).
- Ingestion points: Untrusted data enters the agent context through
client.customers.retrieve,client.customers.list, andclient.customers.wallets.ledgerEntries.listcalls as shown inSKILL.md. - Boundary markers: The provided instructions lack specific delimiters or system prompts to isolate and prevent the agent from obeying instructions that might be embedded within these retrieved fields.
- Capability inventory: The skill enables high-impact operations including customer record modification (
client.customers.update) and the creation of financial ledger entries (client.customers.wallets.ledgerEntries.create), which could be targeted by an injection attack. - Sanitization: There is no evidence of sanitization, filtering, or validation of the data retrieved from the API before it is processed or presented to the agent.
Audit Metadata