finding-healthcare-practices-via-google-search

Warn

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: MEDIUMDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [DYNAMIC_EXECUTION]: The skill includes a customMapFunction parameter that accepts a JavaScript function as a string. This enables the execution of arbitrary code to transform data objects at runtime.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external content from Google Search results, which can be manipulated by third parties.
  • Ingestion points: Data is retrieved from Google Search snippets and titles via the apidojo/google-search-scraper actor referenced in SKILL.md.
  • Boundary markers: There are no defined delimiters or instructions to signal the agent to ignore potentially malicious instructions embedded in the search results.
  • Capability inventory: The skill includes logic to score and rank results based on the content of the retrieved snippets.
  • Sanitization: There is no mention of sanitizing or escaping the text strings returned by the search engine before they are processed by the scoring logic.
  • [COMMAND_EXECUTION]: The skill's documentation provides specific shell command examples using node to execute local scripts (scripts/run_actor.js) and curl to interact with the Apify REST API.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Sep 2, 2026, 07:14 PM
Security Audit — agent-trust-hub — finding-healthcare-practices-via-google-search