modeio-redact
Pass
Audited by Gen Agent Trust Hub on Mar 6, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides PII anonymization and de-anonymization functionality. It accesses user-provided text or files and performs redaction either locally using regex patterns or by sending the content to a vendor-controlled API (safety-cf.modeio.ai). This behavior is documented and aligns with the skill's primary purpose.
- [SAFE]: Sensitive de-anonymization keys (maps) are stored locally in the user's home directory (~/.modeio/redact/maps/) with restrictive file permissions (mode 0o600). The storage logic includes automatic pruning of old maps after 7 days.
- [SAFE]: The skill integrates with established third-party libraries like python-docx and PyMuPDF for document processing. No unauthorized command execution, privilege escalation, or persistence mechanisms were found.
Audit Metadata