non-autoregressive-writing-pass
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill provides structured writing instructions and does not contain any attempts to override safety protocols or system prompts.- [DATA_EXFILTRATION]: There are no network requests, credential usage, or sensitive file path access.- [COMMAND_EXECUTION]: No shell commands, subprocess calls, or script executions are present in the skill files.- [INDIRECT_PROMPT_INJECTION]: The skill processes user drafts for editing purposes. The execution environment described lacks capabilities (such as network access or file system modification) that could be leveraged by adversarial content within a draft.
Audit Metadata