email-campaign
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill incorporates a standard attack surface for indirect prompt injection by fetching data from external websites.
- Ingestion points: The
SKILL.mdfile instructs the agent to fetch and read the user's website and other public sources to establish positioning, audience profiles (ICP), and brand voice. - Boundary markers: The instructions do not specify the use of delimiters or 'ignore instructions' markers when processing the content retrieved from external URLs.
- Capability inventory: The skill involves network fetching and file-write operations to create and update the
.agents/product-context.mdfile, which is then used as a primary source of truth for generating marketing copy. - Sanitization: No explicit sanitization, filtering, or validation process is defined for the content fetched from external websites before it is persisted in the local context file.
Audit Metadata