content-research-writer
Pass
Audited by Gen Agent Trust Hub on Mar 18, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill acts as a collaborative writing partner, offering prompts for outlining, research, and feedback.
- [PROMPT_INJECTION]: No evidence of instructions designed to bypass safety filters or override agent behavior was found.
- [DATA_EXFILTRATION]: The skill does not access sensitive files or perform unauthorized network operations. It suggests local folder creation for document organization which is standard practice.
- [REMOTE_CODE_EXECUTION]: No remote scripts or external dependencies are downloaded or executed.
- [COMMAND_EXECUTION]: Commands included in the documentation (mkdir, cd, touch) are benign and intended for basic file management in the user's local environment.
Audit Metadata