extracting-youtube-comments-for-research

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFEDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [DYNAMIC_EXECUTION]: The skill defines a customMapFunction parameter, which allows passing a JavaScript function string for data transformation. This provides a mechanism for dynamic code execution within the scraper's runtime environment.- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted external data from YouTube comments, creating an attack surface for indirect prompt injection.
  • Ingestion points: Data is ingested from external YouTube video comments via the Apify platform.
  • Boundary markers: There are no explicit instructions for the agent to treat ingested comment text as untrusted data or to use specific delimiters during analysis.
  • Capability inventory: The skill has the ability to execute local scripts via node, perform network operations via curl, and write data to the filesystem using command flags.
  • Sanitization: The skill describes cleaning procedures for length and spam filtering, but lacks robust sanitization to prevent the agent from potentially following instructions embedded within the comments themselves.- [EXTERNAL_DOWNLOADS]: The skill depends on an external service (Apify) and a specific actor (apidojo~youtube-comments-scraper). It also references a local script file (scripts/run_actor.js) that is not part of the provided instruction set.- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands, specifically using node to run scripts and curl to interact with the Apify REST API, which involves network access and local process execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 06:37 PM
Security Audit — agent-trust-hub — extracting-youtube-comments-for-research