healthcare-providers-enrich
Pass
Audited by Gen Agent Trust Hub on Apr 21, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill guides the user to install and update the '@nimbleway/cli' tool from the NPM registry to facilitate data extraction and search operations.
- [COMMAND_EXECUTION]: Executes various shell commands including 'nimble' for web data operations, 'python3' for date manipulation, and 'jq' for processing structured JSON data.
- [DATA_EXPOSURE]: Provides instructions for managing the 'NIMBLE_API_KEY' by exporting it as an environment variable and recommending its storage in shell profiles (.zshrc or .bashrc) for persistence.
- [PROMPT_INJECTION]: Features an attack surface for indirect prompt injection as it processes untrusted content from external healthcare websites and practitioner directories. * Ingestion points: Input data parsing (Step 1) and web content extraction (Step 4) in 'SKILL.md'. * Boundary markers: Not explicitly implemented in the prompt templates. * Capability inventory: Includes 'Bash' execution, 'Agent' tool for sub-agent spawning, and 'Write' access to the '~/.nimble/' directory. * Sanitization: Relies on the Nimble CLI's extraction formatting and LLM-based field detection without specific sanitization filters.
Audit Metadata