skills/aidotnet/moyucode/file-hasher/Gen Agent Trust Hub

file-hasher

Pass

Audited by Gen Agent Trust Hub on Oct 8, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides examples for executing its own script via the command line (e.g., python scripts/file_hasher.py). These commands are standard for local script execution and do not involve shell injection or elevated privileges.- [EXTERNAL_DOWNLOADS]: The skill references Python's official repository as a source in its metadata and comments. These are informational references to a trusted and well-known source and do not involve downloading or executing remote code.- [DATA_EXPOSURE]: The tool reads file content locally to calculate hashes. This is the primary and intended function of the skill. No sensitive files are targeted by default, and there is no mechanism for data exfiltration.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 8, 2026, 09:41 PM