stelow-product-job-to-be-done

Pass

Audited by Gen Agent Trust Hub on Sep 8, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONTIME_DELAYED_CONDITIONAL
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user input for market definitions, purposes, and job performers, which is interpolated into prompt templates across multiple reference files without sanitization or boundary markers.
  • Ingestion points: User-provided market definitions in references/01-contextual-segmentation.md and purpose/performer variables in references/02-thinking-styles.md.
  • Boundary markers: Absent; the skill uses simple brackets for interpolation.
  • Capability inventory: The skill utilizes the 'question' tool to gather user preferences.
  • Sanitization: No sanitization or validation logic is applied to the input before it is used to generate prompts.
  • [TIME_DELAYED_CONDITIONAL]: The SKILL.md entry instructions include conditional logic that gates specific functionality (standalone vs workflow mode) based on the presence of the environment variables STELOW_WORKFLOW and STELOW_STATE.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 8, 2026, 03:01 AM
Security Audit — agent-trust-hub — stelow-product-job-to-be-done