copy-editing

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted marketing copy provided by users, which serves as a potential vector for indirect prompt injection.
  • Ingestion points: User-supplied text triggers the copy-editing sweeps (SKILL.md).
  • Boundary markers: No explicit delimiters or instructions are provided to the agent to ignore potentially malicious content within the user-provided copy.
  • Capability inventory: The skill uses the LLM to transform text and includes instructions to read local configuration files like .claude/product-marketing-context.md (SKILL.md).
  • Sanitization: The skill does not implement specific sanitization or filtering logic for the input text.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 12:32 PM
Security Audit — agent-trust-hub — copy-editing