avoid-ai-writing
Pass
Audited by Gen Agent Trust Hub on Aug 13, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill inherently processes untrusted data (user-provided text and files). It effectively mitigates indirect prompt injection risks by explicitly instructing the agent to treat the content strictly as text under audit and to flag, rather than follow, any instructions found within the data (e.g., 'ignore the rules above').
- [DATA_EXFILTRATION]: No mechanisms for unauthorized data access or network-based exfiltration were detected. The skill focuses entirely on text transformation within the agent's environment.
- [PROMPT_INJECTION]: The instructions do not contain patterns intended to bypass safety filters, extract system prompts, or override the core behavior of the AI agent.
- [REMOTE_CODE_EXECUTION]: The skill does not download external scripts or packages, nor does it use dynamic execution functions like eval() or exec().
- [COMMAND_EXECUTION]: While the skill can modify files using an 'Edit tool', this is restricted to targeted, minimal changes for the stated purpose of writing improvement and includes safeguards against following malicious commands embedded in the files.
Audit Metadata