humanizer-zh
Pass
Audited by Gen Agent Trust Hub on Apr 10, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns or security vulnerabilities were detected in the skill instructions or metadata.
- [DATA_EXFILTRATION]: The skill does not perform any network operations or external data transmissions.
- [REMOTE_CODE_EXECUTION]: There is no evidence of remote script execution, dynamic code loading, or unauthorized command execution.
- [PROMPT_INJECTION]: The instructions do not contain patterns intended to bypass safety filters or override agent behavior.
- [DATA_EXPOSURE]: No sensitive file paths, environment variables, or hardcoded credentials were detected in the source.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted text data via the Read tool (Ingestion point). Boundary markers and explicit sanitization are absent, but the agent's capabilities (Read, Write, Edit, AskUserQuestion) are appropriately scoped to the text editing task, minimizing the risk of multi-step exploitation.
Audit Metadata