field-service-foundation-setup-designer-get

Pass

Audited by Gen Agent Trust Hub on Sep 7, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill is designed to fetch content from user-provided URLs to extract business context and signals for generating Field Service recommendations.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from multiple sources (URLs, JSON, CSV, and free-form text) which are interpolated into the agent's context to drive a design interview.
  • Ingestion points: The 'Pre-phase: Source ingestion' and 'Inputs' sections of SKILL.md describe fetching user-supplied URLs and accepting pasted data in various formats.
  • Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded commands within the ingested content.
  • Capability inventory: While the skill itself runs a design conversation, it generates JSON meant for deployment via the 'sfs-sobject-create' tool.
  • Sanitization: There is no mention of sanitizing or validating the ingested content before it is used to influence the agent's recommendations.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 7, 2026, 02:30 PM
Security Audit — agent-trust-hub — field-service-foundation-setup-designer-get