copy-editing

Pass

Audited by Gen Agent Trust Hub on Apr 5, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection attack surface because it is designed to ingest and process untrusted text (marketing copy) for editing and refinement without implementing defensive boundary markers.
  • Ingestion points: The skill primarily processes user-provided copy, drafts, and marketing materials for editing as specified in SKILL.md.
  • Boundary markers: Absent. There are no instructions for the agent to use delimiters or specific framing (like XML tags or markdown code blocks) to separate user data from the skill's operational instructions.
  • Capability inventory: The skill is configured to read project-specific context from .claude/product-marketing-context.md and generates modified text outputs.
  • Sanitization: The skill lacks mechanisms to sanitize or validate input text to ensure it does not contain malicious instructions intended to subvert the agent's behavior during the editing process.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 5, 2026, 08:05 PM
Security Audit — agent-trust-hub — copy-editing