kai-brief
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from local project files to generate marketing strategies, creating a surface for instructions in those files to influence agent behavior.
- Ingestion points: Reads
MARKETING.md,README.md,PROJECT.md,package.json, and searches for 'email/ad/analytics config' to build product context. - Boundary markers: No specific delimiters or 'ignore' instructions are provided for the ingested project context to prevent the agent from obeying instructions found within those files.
- Capability inventory: The skill generates a structured brief for downstream writing tasks; it does not request network or command execution permissions.
- Sanitization: Input content from project files is not filtered or sanitized before being used to populate the content brief schema.
Audit Metadata