marketing-council
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input from user queries and external sources, creating a potential vector for indirect prompt injection.\n
- Ingestion points: According to
SKILL.md, the agent reads user-provided marketing questions, external search results from the 'Live Research Pass', and project context files like.agents/product-marketing.md.\n - Boundary markers: The instructions do not specify any delimiters or safety prompts to isolate and ignore potentially malicious instructions embedded within these external context sources.\n
- Capability inventory: The skill includes instructions in
SKILL.mdto read local project files, write custom advisor dossiers to the.agents/advisors/directory, and execute search/research tools.\n - Sanitization: There is no documented evidence of sanitization, filtering, or validation of the data retrieved from user prompts or web sources before it is processed by the agent.
Audit Metadata