product-launch
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill acts as a processing hub for external data, which serves as a potential surface for indirect prompt injection.
- Ingestion points: Ingests user-supplied descriptions and external 'CRM/call signals' as specified in
references/task-method.md. - Boundary markers: The provided instructions do not include explicit delimiters or instructions to ignore embedded commands within ingested signals.
- Capability inventory: The agent is authorized to use tools for record hydration, automated content generation, and 'workspace mutations' for data persistence across
SKILL.mdand its references. - Sanitization: The skill lacks defined validation or escaping mechanisms for external inputs before they are interpolated into generation prompts.
Audit Metadata