drug-labels-search
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests drug label data from api.valyu.ai, which serves as a potential vector for indirect prompt injection if the source data contains adversarial content.
- Ingestion points: Data is fetched via
scripts/search.mjs. - Boundary markers: None are specified for the search result output.
- Capability inventory: Network access for API calls and file system access for API key configuration.
- Sanitization: Content is returned as a JSON structure without explicit filtering of text fields.
- [COMMAND_EXECUTION]: The documentation provides a shell command sequence to dynamically resolve the script path in the user environment.
Audit Metadata