marketing-mix-modelling
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it processes data from external sources that could contain adversarial content.
- Ingestion points: The workflow ingests weekly outcome series from client systems (POS, CRM), media spend and exposure data from agency exports, and external events from public calendars (
SKILL.md,references/data-readiness-and-specification.md). - Boundary markers: The skill implements pre-defined validation and reading rules (Workflow Step 7) and quality standards (Quality Standards section) to gate the interpretation of results.
- Capability inventory: The skill's primary function is to generate text deliverables such as model specifications and calibration plans. It explicitly states that analysis is read-only and that executing code or modifying live accounts requires separate, specific authority (
SKILL.md). - Sanitization: The skill includes a quality review step (Workflow Step 9) to verify controls, ROI plausibility, and the sourcing of priors before hand-over.
Audit Metadata