author
Pass
Audited by Gen Agent Trust Hub on Jun 19, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a mandatory 'Draft-first rule' and 'Draft Approval Gate.' This ensures that no content is written to the file system without explicit user review and approval of the full proposed text.
- [SAFE]: The style guide (reference/style.md) enforces a strict ban on typographic and 'smart' characters, requiring standard ASCII only. This reduces the risk of homoglyph attacks or hidden characters being introduced into the workspace.
- [SAFE]: The skill requires a structured research phase where all factual claims must be supported by verifiable source URLs. This allows users to audit the origins of information processed by the agent.
- [SAFE]: The workflow requires reading the full article before any edits are made. This 'whole-article awareness' helps ensure consistency and prevents an attacker from injecting fragmented instructions that might be missed in an isolated edit.
Audit Metadata