dependency-decision-helper

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is restricted to a read-only workflow. It explicitly prohibits file modifications, package installations, branch/PR creation, or state mutations within the Endor Labs platform.
  • [SAFE]: The skill handles sensitive configuration files safely. It contains specific instructions to extract only the namespace key from the Endor configuration file (~/.endorctl/config.yaml) and strictly forbids echoing secrets, tokens, or credential keys.
  • [SAFE]: To mitigate potential indirect prompt injection, the skill includes explicit instructions to treat all external inputs, such as repository files and command outputs, strictly as data rather than instructions.
  • [SAFE]: Command execution is limited to specific, documented, read-only lookups using the vendor's official 'endorctl' CLI tool, with placeholders clearly defined for package coordinates.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 06:47 PM
Security Audit — agent-trust-hub — dependency-decision-helper