ears-authoring
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOW
Full Analysis
- [PROMPT_INJECTION] (SAFE): No instructions to override behavior or bypass safety constraints were found.
- [DATA_EXFILTRATION] (SAFE): No network operations or access to sensitive credentials or file paths are present.
- [REMOTE_CODE_EXECUTION] (SAFE): The skill does not download or execute remote scripts or external packages.
- [COMMAND_EXECUTION] (SAFE): No arbitrary shell commands or subprocess spawning are used.
- [INDIRECT_PROMPT_INJECTION] (SAFE): The skill processes user-provided informal requirements for formatting purposes. There is no automated execution logic or high-privilege side effects from the processed data.
Audit Metadata