copywriting
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input to perform copywriting tasks.\n- Ingestion points: Untrusted data enters the agent's context through user requests to write, rewrite, or improve marketing copy, as well as via the
.claude/product-marketing-context.mdfile.\n- Boundary markers: No specific delimiters or instructions to ignore embedded commands are used to wrap the processed content.\n- Capability inventory: The skill utilizes the agent's natural language generation and file reading. No evidence of shell command execution, network requests, or sensitive file access was found in the provided files.\n- Sanitization: The skill lacks mechanisms to sanitize or escape user-provided text before processing it.
Audit Metadata