field-service-work-rule-designer-configure

Pass

Audited by Gen Agent Trust Hub on Sep 7, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill facilitates a structured interview to collect configuration parameters, such as field names and business constraints, directly from user responses (SKILL.md). These inputs are bundled into a workRules[] JSON block and passed to the next agent in the sequence (sfs-service-objective-designer).
  • Ingestion points: User-provided strings for field names (serviceProperty, resourceProperty, booleanField), numeric values, and descriptive logic.
  • Boundary markers: None present; the instructions do not specify delimiters or 'ignore' instructions for the interpolated user content.
  • Capability inventory: The skill itself has no shell, file, or network capabilities, but it functions as a primary data source for subsequent agents that may have such privileges.
  • Sanitization: Absent; instructions explicitly state to 'pass through as given' for field names provided by the user, creating a risk if downstream handlers do not perform validation.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 7, 2026, 02:34 PM
Security Audit — agent-trust-hub — field-service-work-rule-designer-configure