kpic-drug-search
Pass
Audited by Gen Agent Trust Hub on Nov 2, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill utilizes the @nomadamas/k-skill CLI via npx to manage instructions and updates. These are vendor-owned packages provided by the skill's author.
- [COMMAND_EXECUTION]: The skill executes a local Node.js helper script to perform network requests to official drug information domains.
- [INDIRECT_PROMPT_INJECTION]: The skill processes pharmaceutical data from external APIs. It provides specific instructions to the AI to maintain the distinction between raw data and summaries, mitigating the risk of following instructions potentially embedded in web content.
- [SAFE]: Explicit 'Hard rules' are defined to prevent the AI from bypassing security boundaries, storing credentials, or performing sensitive actions without user consent.
Audit Metadata