Evals Integration

Spec-to-Evals Generator

Generate contract validity, breaking-change, MCP schema, webhook, OpenAPI conformance, AgentUI render, Sandbox, BrowserOps, Monitor, and policy eval suites.

Generate Evals Suite

Public Suite Drafts

Generated eval suites are local artifacts with source hashes and limitations.

contract validity eval

breaking change eval

webhook contract eval

policy compliance eval

Protected Trigger

Remote Evals execution requires PLATPHORM_API_KEY and returns real evidence only when the downstream service responds.

trigger_eval MCP tool

release gate evidence

scorecard handoff