Eval gate
An eval gate is a CI quality gate that re-tests an AI agent against its versioned evaluation dataset on every change — to the agent, its skills or its tool bundles — and compares the result to a committed baseline score. Regressions are flagged before the change reaches production.
In NeoXam Agents
Every agent ships with a Git-versioned eval dataset and a baseline score that can only be raised. CI re-runs evals on every descriptor, skill or bundle change — advisory during the Beta, blocking at GA. The working principle: an agent that cannot be evaluated cannot be trusted.
Related terms
See it at work
Business rule validation & testing
Shipped (V0)On every rule save, DataHub chains governed agents: validate syntax and best practice, evaluate against business assertions, build and run tests.
See how these concepts come together in a governed agentic platform for investment operations.
Explore the platform