Version updated for https://github.com/seva9523/EvalRepro to version v0.1.0a2. This action is used across all versions by ? repositories. Action Type This is a Composite action. Go to the GitHub Marketplace to find the latest changes. Action Summary EvalRepro is an open-source tool designed to detect semantic drift in AI evaluation inputs and contracts before changes can affect benchmark results silently. It creates hash-only manifests in isolated environments and compares the evaluation contract rather than relying solely on import success. The action checks various aspects such as task version, adapter parameters, sample coverage, order, and semantic fields, providing verdicts like reproducible, order_drift, or semantic_drift.