monitoring-google-news-coverage-for-brand

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests data from Google News search results. Adversarially crafted headlines or snippets could potentially influence the agent's logic during classification or reporting steps.\n
  • Ingestion points: Google News search results (headlines, snippets, and publications) fetched via the Google Search scraper actor.\n
  • Boundary markers: No delimiters or specific instructions to ignore embedded commands are provided.\n
  • Capability inventory: Execution of shell commands (node scripts/run_actor.js), network requests (curl), and file system writes (--output flag).\n
  • Sanitization: The skill does not mention explicit sanitization or filtering for external content.\n- [DYNAMIC_EXECUTION]: The customMapFunction input parameter allows for a JavaScript function to be passed as a string to be executed for data transformation on the Apify platform.\n- [COMMAND_EXECUTION]: The skill provides explicit shell commands for the agent to execute, including running a local Node.js script and making API calls via curl.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 06:52 PM
Security Audit — agent-trust-hub — monitoring-google-news-coverage-for-brand