scorable-integration
Pass
Audited by Gen Agent Trust Hub on May 5, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides integration instructions for Scorable, a legitimate tool for LLM-as-a-Judge evaluation.
- [SAFE]: The skill promotes secure credential management by instructing users to use environment variables and cautioning against sharing keys in the prompt.
- [SAFE]: All identified external resources, including software packages (scorable, @root-signals/scorable) and API endpoints (api.scorable.ai), are official assets of the vendor.
- [SAFE]: Indirect Prompt Injection Surface: The skill analyzes user-provided codebase files to identify LLM interactions. -- Ingestion points: Codebase files analyzed during the assessment phase. -- Boundary markers: None specified; relies on the agent's contextual awareness. -- Capability inventory: Writing integration code, installing packages via standard managers, and making API requests to Scorable's domain. -- Sanitization: Relies on the agent's internal safety filters for code generation and analysis.
Audit Metadata