ae-metadata

Pass

Audited by Gen Agent Trust Hub on Aug 21, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [SAFE]: The skill serves as a comprehensive management interface for a metadata capability gateway. It adheres to safety protocols by enforcing the use of --dry-run and --validate flags for complex operations and requiring explicit confirmation for high-risk write operations like table deletion.
  • [COMMAND_EXECUTION]: The skill uses the ae-cli tool to perform various metadata tasks. All commands are well-defined, and the skill provides clear rules for discovering and verifying resource IDs (e.g., project_id, data_table_id) rather than allowing the agent to invent them. The host override capability is a documented feature for switching environment contexts and is not used for unauthorized exfiltration in the provided instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 21, 2026, 06:39 AM
Security Audit — agent-trust-hub — ae-metadata