competitors
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to perform research by ingesting data from external sources and local configuration files, which presents an attack surface for indirect instructions.
- Ingestion points: The skill reads product marketing context from files like
.agents/product-marketing.mdand performs 'review mining' from external sites such as G2, Capterra, and TrustRadius. - Boundary markers: The instructions do not define specific delimiters or security headers to separate untrusted research data from the agent's primary instructions.
- Capability inventory: The skill generates structured YAML data and markdown content for marketing pages.
- Sanitization: There is no requirement for the agent to sanitize or validate the content retrieved from external competitor reviews or feedback sessions.
Audit Metadata