final
Pass
Audited by Gen Agent Trust Hub on Apr 2, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill contains only natural language instructions and formatting guidelines. It does not include executable code, scripts, or binary files.
- [DATA_EXFILTRATION]: While the skill defines a persistence layer at
.sumeru/finalize/, this is intended for local storage of editing reports and formatted chapters. There are no network calls or instructions to move data outside the local environment. - [COMMAND_EXECUTION]: The skill mentions features like 'batch replacement' and 'automatic segmentation.' These are described as logic for the agent to apply during text processing and do not involve the execution of arbitrary shell commands.
- [PROMPT_INJECTION]: The skill does not contain instructions designed to bypass agent safety filters or override system-level constraints. It strictly focuses on novel auditing and publishing preparation.
Audit Metadata