managing-dagster

Pass

Audited by Gen Agent Trust Hub on Jun 17, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill uses environment variables (DAGSTER_API_TOKEN, DAGSTER_URL) to handle credentials and connection endpoints, which prevents hardcoding secrets and adheres to standard security practices.
  • [SAFE]: All network operations are directed to the user-configured Dagster instance for legitimate GraphQL API interactions.
  • [SAFE]: The skill includes robust 'Discovery-First' and 'Anti-Hallucination' rules that force the agent to verify the existence of resources before performing actions, reducing the risk of operational errors.
  • [SAFE]: Analysis of the provided shell scripts and instructions revealed no signs of obfuscation, malicious persistence mechanisms, or unauthorized privilege escalation.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 17, 2026, 11:28 AM
Security Audit — agent-trust-hub — managing-dagster