copy-editing

Pass

Audited by Gen Agent Trust Hub on Mar 5, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is designed to process and 'sweep' user-provided marketing copy, which constitutes a surface for indirect prompt injection. Malicious instructions could be embedded in the text being edited to attempt to influence the agent's behavior.
  • Ingestion points: The agent accepts user-provided text for editing and reads the local file .agents/product-marketing-context.md (or .claude/product-marketing-context.md).
  • Boundary markers: The instructions do not define explicit delimiters or 'ignore' instructions for the content being processed.
  • Capability inventory: The skill contains no scripts, subprocess calls, file-write operations, or network requests.
  • Sanitization: No input sanitization or validation logic is present in the markdown instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 5, 2026, 09:36 AM