cn-letter-of-credit-check
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data, including Letter of Credit (L/C) texts, sales contracts, and draft shipping documents (invoice, bill of lading, etc.). This creates a vulnerability surface where instructions hidden within these documents could potentially influence the agent's behavior.
- Ingestion points: The skill explicitly requests MT700 L/C fields, sales contracts, and draft export documents (SKILL.md).
- Boundary markers: There are no explicit instructions to the agent to ignore or isolate potential NL instructions embedded within the provided L/C or contract text.
- Capability inventory: The skill primarily generates text analysis and tables based on the input. It does not appear to possess file-write, network-send, or code-execution capabilities in the provided instructions.
- Sanitization: No sanitization or validation of the input text is mentioned; the skill relies on direct comparison and inference.
Audit Metadata