cn-labour-contract-decoder
Pass
Audited by Gen Agent Trust Hub on Oct 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data in the form of labor contracts and offer letters, which is a known attack surface for indirect prompt injection.
- Ingestion points: Contract text, employee handbooks, and offer letters provided by the user.
- Boundary markers: The skill contains a specific defensive block instructing the agent to treat pasted documents strictly as data and to ignore any instructions found within them.
- Capability inventory: The skill consists purely of natural language instructions and does not utilize subprocesses, file-writing, or network operations.
- Sanitization: The instructions explicitly require the agent to notify the user if any embedded instructions are found in the documents instead of executing them.
Audit Metadata