ljg-classic

Pass

Audited by Gen Agent Trust Hub on Oct 3, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local TypeScript scripts using the Bun runtime (Tools/RenderClassic.ts and Tools/ValidateClassic.ts) to handle the rendering and validation logic. This is the primary function of the skill and is performed within a restricted environment.
  • [EXTERNAL_DOWNLOADS]: The workflow involves installing the playwright package and its Chromium browser binaries. These are well-known, established tools from a trusted organization (Microsoft) and are necessary for the skill's rendering capabilities.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied classical Chinese text to generate annotations and interpretations. While this presents an attack surface where malicious instructions could be embedded in the text, the skill mitigates this by applying HTML escaping (escapeHtml) before rendering the content in a headless browser and by using a strictly defined JSON schema for input.
  • [DYNAMIC_EXECUTION]: The skill dynamically generates HTML templates at runtime, which are then rendered into images via a headless browser (Playwright). The renderer includes safeguards such as rejecting remote image URLs and performing DOM audits to ensure the generated content is consistent and safe.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 3, 2026, 09:12 AM