architect-migrate-from-competitor
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is designed to process untrusted external data in the form of agent configuration exports (JSON/YAML) from platforms like Retell, Vapi, or Bland. It includes robust security instructions to audit this data and explicitly forbids storing credentials in plain text, requiring the use of the platform's secure secrets management API (
/v1/convai/secrets) instead. - [SAFE]: The instructions for verifying third-party endpoints mentioned in the agent prompt include specific safety constraints: the agent must obtain user consent, use only read-only endpoints, use synthetic values for probes, and must not use credentials extracted from the export for these exploratory calls.
- [SAFE]: The skill follows a delegation model, routing specific technical tasks (like security auditing, model selection, and tool creation) to specialized sub-skills, ensuring that the overall migration process adheres to established safety and architectural guidelines.
- [SAFE]: No evidence of prompt injection, obfuscation, unauthorized persistence, or malicious code execution was found. The skill operates within the expected bounds of a migration and configuration tool.
Audit Metadata