cognitive-output-refiner
Pass
Audited by Gen Agent Trust Hub on Jul 4, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements robust data protection rules. It explicitly requires the redaction of secrets, API keys, credentials, and private identifiers (like local usernames and internal hostnames) from its output, prioritizing security over exact preservation.
- [SAFE]: The skill processes external data such as logs and console outputs, which is an inherent indirect prompt injection surface. It mitigates this risk by providing strict instructions on fidelity and redaction, and it lacks dangerous capabilities like network access or arbitrary command execution.
- [COMMAND_EXECUTION]: No shell commands, subprocess calls, or system-level execution patterns were detected.
- [DATA_EXFILTRATION]: No network operations (e.g., curl, fetch) or mechanisms for sending data to external domains are present.
- [REMOTE_CODE_EXECUTION]: The skill does not download or execute external scripts or packages.
Audit Metadata