assemblyai-core-workflow-a
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements strong security boundaries by requiring human approval for live production actions, including credential changes, deployment, and data deletion.
- [SAFE]: The instructions mandate secure credential handling, explicitly forbidding the printing, committing, or exposure of the API key, and recommending the use of approved secret managers.
- [SAFE]: The skill references official documentation from AssemblyAI, a well-known technology service, to guide implementation without downloading external code or scripts.
- [SAFE]: The workflow specifically instructs the agent to exclude credentials, PII, and customer-derived content from its outputs to prevent accidental data exposure.
Audit Metadata