long-image

Warn

Audited by Gen Agent Trust Hub on Nov 12, 2026

Risk Level: MEDIUMDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [DYNAMIC_EXECUTION]: The skill scripts scripts/doctor.mjs and scripts/long.mjs use dynamic imports to load core rendering logic from a path computed at runtime. The target path is determined by the ARTICLE2MEDIA_ENGINE environment variable or a relative directory path (../../article2media/engine/scripts/). This allows for the loading and execution of code from a dynamically computed location.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted content from user-provided files (article.md and content.json) to generate visual media. Ingestion points: Reads external files article.md and content.json. Boundary markers: The skill does not provide instructions or delimiters to isolate untrusted content from agent commands. Capability inventory: The skill has access to the filesystem, shell command execution, and dynamic module loading. Sanitization: There is no evidence of sanitization or validation of the input content within the provided skill scripts.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Nov 12, 2026, 04:49 PM