b2c-checkout-triage
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external sources, which could potentially contain malicious instructions intended to influence the agent's behavior.
- Ingestion points: Application logs accessed via
logs_get_recent, order records via Admin Orders contracts, and product/attribute data via SCAPI (found in SKILL.md). - Boundary markers: The skill provides instructions to "Avoid customer/payment data in output" and to preserve query errors, but lacks explicit prompt delimiters or "ignore instructions" wrappers for the ingested data.
- Capability inventory: The skill uses specialized tools (
b2cCLI) and API reads for diagnostic purposes; it does not include instructions for file-write operations, persistent storage modification, or arbitrary network exfiltration. - Sanitization: No specific sanitization or filtering of log content or record data is mentioned prior to processing.
Audit Metadata