tajemnica-preflight-pl
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill operates entirely locally as per its configuration (
data-residency: local) and explicit instructions. It serves as a safety filter to prevent accidental data leaks by human users. - [SAFE]: The skill includes a 'requires-human-approval: true' flag and explicitly states in its instructions that it does not send data anywhere and requires the user to manually handle the final output.
- [SAFE]: All identified legal references and logic relate to Polish and EU law (RODO/GDPR, Attorney Privilege) and are used for their intended purpose of risk assessment.
- [SAFE]: No obfuscation, prompt injection, or malicious command execution patterns were detected. The skill uses standard placeholders like
[KLIENT]for redaction drafts, which is a best practice for privacy.
Audit Metadata