occupational-health-analyzer
Pass
Audited by Gen Agent Trust Hub on Sep 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from various local files, including 'occupational-health-tracker.json', 'sleep-tracker.json', and 'mental-health-tracker.json'.
- Ingestion points: Data is read during 'Step 1: Data Reading' and the '关联分析' (Association Analysis) phase.
- Boundary markers: The instructions do not define specific delimiters or instructions for the agent to ignore embedded commands within the JSON data.
- Capability inventory: The skill uses 'Read', 'Write', and 'Edit' tools to interact with the file system.
- Sanitization: No explicit sanitization or validation logic is provided to filter malicious instructions that might be embedded in the health tracking data.
- [DATA_EXPOSURE]: The skill is designed to handle sensitive personal health information. While it lacks network capabilities to exfiltrate this data, it has permissions to read and modify files, creating a potential exposure surface if the agent context is compromised.
Audit Metadata