skill-engineering

Pass

Audited by Gen Agent Trust Hub on Jul 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [SAFE]: No malicious behavior, prompt injection, or obfuscation was detected. The skill is instructional and methodological, focusing on improving agent judgment and action through structured review.- [EXTERNAL_DOWNLOADS]: The refresh-sequence command enables fetching updated principle interpretations from the vendor's official GitHub repository (github.com/lidessen/skills.git). This is a legitimate maintenance function that includes explicit validation of the source data structure to prevent malformed or unauthorized content injection.- [COMMAND_EXECUTION]: The sync-sequence-snapshot command provides instructions for maintainers to execute a local Python script (scripts/sync-sequence-snapshot.py) to update packaged snapshots of principles. This is a standard development workflow and does not involve arbitrary or unsafe command execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 17, 2026, 06:31 AM
Security Audit — agent-trust-hub — skill-engineering