skills/huangzida/skills/vitepress/Gen Agent Trust Hub

vitepress

Pass

Audited by Gen Agent Trust Hub on Oct 6, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill describes features for Data Loaders and Dynamic Routes that ingest data from external APIs or local files, which are then processed in a build-time Node.js environment.
  • Ingestion points: references/features-data-loading.md and references/features-dynamic-routes.md provide examples of fetching data from remote URLs via fetch() and reading the local file system using node:fs.
  • Boundary markers: The provided code examples do not include explicit sanitization or boundary markers for the ingested external data.
  • Capability inventory: Documentation across various files (e.g., references/core-config.md, references/features-data-loading.md) details capabilities including network access, file system read/write operations, and execution of arbitrary JavaScript/TypeScript code during the build process.
  • Sanitization: The examples demonstrate direct processing of external content (such as JSON.parse or rendering raw Markdown/HTML) without explicit validation or filtering logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 6, 2026, 04:20 PM