humanize
Pass
Audited by Gen Agent Trust Hub on Aug 11, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists entirely of natural language instructions and examples for text editing. It does not use any tools, execute shell commands, or perform network requests.
- [DATA_EXPOSURE]: The skill suggests reading an optional
ABOUT-ME.mdfile to personalize the writing style to the user. This is a standard practice for context-aware agents and occurs within the local environment without external transmission. - [PROMPT_INJECTION]: The instructions contain phrases like "You are a writing editor" and "If ABOUT-ME.md exists... read the Communication Style section," which are standard role-setting and conditional instructions for an AI agent. They do not attempt to bypass safety filters or override system constraints.
- [EXTERNAL_DOWNLOADS]: The skill includes a reference link to a well-known service (Wikipedia). This is a safe reference for documentation purposes and does not involve downloading executable content.
Audit Metadata