collection-filter-authoring
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: No attempts to override safety guidelines or agent behavior were detected. Instructions focus on technical authoring tasks for the Datex platform.
- [DATA_EXPOSURE]: No sensitive data exposure or exfiltration patterns were identified. The skill uses standard CLI commands for legitimate metadata exploration and constant discovery.
- [OBFUSCATION]: No obfuscated code, encoded strings, or hidden content were found in the instructions or reference files.
- [REMOTE_CODE_EXECUTION]: No remote code execution patterns or untrusted script downloads were detected. External links point to the vendor's official GitHub repository for documentation and reference.
- [DYNAMIC_CONTEXT_INJECTION]: The skill does not use dynamic shell execution patterns (e.g., Load-time command execution) in its configuration or documentation.
Audit Metadata