ace-step
Pass
Audited by Gen Agent Trust Hub on May 18, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill uses the official @runcomfy/cli package from the NPM registry. This is a trusted vendor resource from agentspace-so that facilitates interaction with their model APIs.
- [SAFE]: API token management follows secure standards, directing users to utilize environment variables or configuration files with restricted file permissions (0600) to prevent unauthorized access.
- [SAFE]: Command execution is handled via structured JSON input passed to the CLI, which serves as a boundary marker to prevent shell injection from prompt content.
- [SAFE]: The skill explicitly documents the risk of indirect prompt injection from untrusted audio URLs and provides specific agent mitigations, such as only processing user-provided URLs and monitoring for output divergence.
- [SAFE]: Instructions include clear security and privacy sections that forbid the use of arbitrary remote install scripts, ensuring that installation is performed only through verified package managers.
Audit Metadata