autoevals
Universal library for evaluating AI models
This report shows the evidence for autoevals@0.5.0. 1 of 2 known versions is blocked or flagged.
AI security review
scanned · by lpm-firewall-aiThe default evaluation path can send OPENAI_API_KEY, an OpenAI credential, to the package publisher’s Braintrust gateway. This is a concrete credential exposure in the product’s visible request path, so it warrants a block under the credential rule.