copy-editing
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill instructions and associated reference files do not contain any malicious code, external dependencies, or network-based exfiltration patterns.
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the processing of user-supplied text for copy editing, which is an untrusted data ingestion point. The use of a structured seven-sweep framework guides the agent's behavior, though the instructions do not include explicit boundary markers or sanitization steps for the input copy.
- Ingestion points: Marketing copy provided by the user for review (SKILL.md).
- Boundary markers: Absent.
- Capability inventory: The agent reads local context from .claude/product-marketing-context.md (SKILL.md).
- Sanitization: No input validation or filtering is defined for the content being processed.
Audit Metadata