Skills
skills/lightfastai/lightfast/router-core/data-loading/Gen Agent Trust Hub

router-core/data-loading

Pass

Audited by Gen Agent Trust Hub on Jun 6, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill documents the ingestion and processing of untrusted data from URL path parameters and search parameters (e.g., postId, offset, limit). This creates a standard surface for indirect prompt injection common in routing applications.
  • Ingestion points: External data enters the context through params and search objects in SKILL.md.
  • Boundary markers: The documentation does not specify explicit delimiters for the AI to distinguish between data and instructions when processing these parameters.
  • Capability inventory: The skill demonstrates the use of the fetch API to perform network operations based on the ingested parameters.
  • Sanitization: The provided code examples include basic input validation and sanitization, such as converting search parameters to numbers using Number(search.offset) || 0 and validating search schemas via validateSearch.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 6, 2026, 03:41 AM
Security Audit — agent-trust-hub — router-core/data-loading