webinar-plan
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes brand data and agency SOPs from the local file system to inform its generation process.
- Ingestion points: Loads JSON and Markdown files from the
~/.claude-marketing/directory andskills/context-engine/compliance-rules.md. - Boundary markers: The skill does not define specific delimiters to separate the loaded brand context from the agent's instructions.
- Capability inventory: The skill utilizes text-based agents (content-creator, email-specialist, marketing-strategist) and has no documented access to network tools, system shells, or file writing capabilities.
- Sanitization: No data validation or sanitization steps are defined for the imported brand profiles or guidelines.
Audit Metadata