ai-automation-workflows
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFE
Full Analysis
- [SAFE] The skill documents the use of the
beltCLI tool for AI orchestration and provides various shell and Python script templates for common automation tasks. - [SAFE] Includes standard automation patterns such as rate limiting (
sleep), retries with exponential backoff, and conditional logic based on model outputs. - [SAFE] Demonstrates integration with external webhooks for alerting using placeholder URLs (
your-webhook.com). - [SAFE] References the installation of related skills from the
inference-shandbelt-shorganizations, which are the providers of the tools described. - [SAFE] The scripts perform local file operations (creating output directories and JSON files) and standard network requests to the vendor's API via the CLI tool, with no unauthorized access or exfiltration detected.
Audit Metadata