stop-slop
Pass
Audited by Gen Agent Trust Hub on Oct 18, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill operates entirely through instructional prompts and reference markdown files. It does not include scripts, executables, or commands that interact with the host system.
- [SAFE]: All referenced domains (e.g., Acme, LinkedIn, social platforms) are used within the context of illustrative writing examples and do not involve network exfiltration or credential harvesting.
- [SAFE]: The 'matching a voice' feature instructs the agent to ask the user for writing samples rather than scraping external URLs or accessing unauthorized data sources.
- [SAFE]: Instructions for 'placeholders' (e.g.,
[how many customers signed up]) are used for legitimate editorial purposes to identify where a user needs to provide specific facts, rather than as a mechanism for data exfiltration. - [SAFE]: The skill's frontmatter correctly limits its scope to text humanization and provides detailed guidelines on maintaining factual integrity during the rewriting process.
Audit Metadata