schema
Pass
Audited by Gen Agent Trust Hub on Aug 22, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions direct the agent to read context from workspace files like
.agents/product-marketing.mdduring its initial assessment phase. This introduces an attack surface where data from external files is ingested into the agent's context, potentially influencing its behavior if the files are compromised. - Ingestion points:
.agents/product-marketing.md,.claude/product-marketing.md, andproduct-marketing-context.md(referenced in the Initial Assessment section of SKILL.md). - Boundary markers: None identified for distinguishing context data from instructions.
- Capability inventory: The skill utilizes file reading and text generation.
- Sanitization: No validation or sanitization of the content read from context files is specified.
- [EXTERNAL_DOWNLOADS]: The skill references official validation tools from Google (Rich Results Test) and the Schema.org Validator. These are recognized as well-known, trusted services for verifying structured data implementation.
Audit Metadata