copy-editing
Pass
Audited by Gen Agent Trust Hub on Aug 26, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides a structured framework and checklists for copy editing marketing text. It is entirely composed of markdown files and does not contain scripts, external dependencies, or network-enabled commands.\n- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process user-provided marketing copy. Ingestion point: user input provided to the agent for editing (SKILL.md). Boundary markers: none. Capability inventory: none (no scripts or command execution instructions). Sanitization: none. The absence of dangerous capabilities, such as shell access or network requests, ensures that the data ingestion surface remains safe.
Audit Metadata