researching-internet-slang-and-cultural-trends

Warn

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: MEDIUMDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [DYNAMIC_EXECUTION]: The skill defines a customMapFunction parameter described as a 'JavaScript function to transform each output object'. This allows the execution of arbitrary JavaScript code provided as a string at runtime, which is a common vector for code injection attacks.\n- [INDIRECT_PROMPT_INJECTION]: The skill ingests user-generated content from Urban Dictionary, a crowdsourced platform where entries are untrusted and can be manipulated to include hidden instructions or harmful content.\n
  • Ingestion points: Slang definitions, usage examples, and metadata fetched from Urban Dictionary via the scraper.\n
  • Boundary markers: None identified in the provided skill instructions.\n
  • Capability inventory: File writing (via --output flag in CLI examples) and network communication (via Apify API).\n
  • Sanitization: No evidence of sanitization or filtering of the external content before processing or display.\n- [EXTERNAL_DOWNLOADS]: The skill utilizes the apidojo/urbandictionary-scraper actor on Apify. While Apify is a well-known service and the resource is owned by the skill author, it represents a dependency on external code executed in a remote environment.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Sep 2, 2026, 06:31 PM
Security Audit — agent-trust-hub — researching-internet-slang-and-cultural-trends