copy-editing
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process marketing copy from user input and read from the .claude/product-marketing-context.md file, creating an attack surface for indirect instructions.
- Ingestion points: Marketing copy provided for review and the content of the .claude/product-marketing-context.md context file.
- Boundary markers: The instructions do not define explicit boundary markers or delimiters to isolate the ingested copy from the agent's internal instructions.
- Capability inventory: The skill instructs the agent to read local files to gather brand and marketing context.
- Sanitization: There are no mentioned mechanisms for sanitizing or escaping the content of the ingested text before the editing sweeps are performed.
Audit Metadata