copy-editing
Pass
Audited by Gen Agent Trust Hub on Jun 21, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns or security risks were identified in the analysis. The skill functions as a procedural guide for text manipulation.
- [DATA_EXPOSURE]: The skill references standard local context files such as
.agents/product-marketing.md. This is a best-practice pattern for context-awareness and does not involve unauthorized data access or exfiltration. - [PROMPT_INJECTION]: The instructions are strictly limited to the domain of copy editing and conversion optimization. There are no attempts to override system prompts, bypass safety filters, or extract underlying model configurations.
- [REMOTE_CODE_EXECUTION]: There is no evidence of external downloads, script execution, or package installations. All provided logic resides within the static markdown and documentation files.
- [COMMAND_EXECUTION]: The skill does not utilize shell commands or subprocess calls. Its primary function is to transform and critique natural language provided by the user.
Audit Metadata