finding-food-bloggers-and-creators

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONDYNAMIC_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from Instagram, TikTok, and YouTube (bios, posts). There are no explicit boundary markers or sanitization steps mentioned to prevent this data from influencing agent instructions. Ingestion points: Social media profile data and post content scraped via Apify actors. Boundary markers: Absent. Capability inventory: Network requests (curl) and script execution (node). Sanitization: Not explicitly defined for the ingested text.
  • [COMMAND_EXECUTION]: The skill instructs the agent to use curl for API fallback and node to run a local script (scripts/run_actor.js). These commands use environment variables for authentication and could be misused if command parameters are manipulated.
  • [DYNAMIC_EXECUTION]: The customMapFunction input allows for a JavaScript function string to be passed to the Apify platform for data transformation. While a standard feature of the underlying service, it involves dynamic code execution based on provided input.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 08:12 PM
Security Audit — agent-trust-hub — finding-food-bloggers-and-creators