Run Scale AI’s AgentEnv evals on Vercel Sandbox

Scale AI's AgentEnv agent-evaluation framework now runs on Vercel Sandbox, with the Vercel provider shipping in AgentEnv 0.9.1334 and later and running the agent and its task environment in separate Sandboxes. Vercel published an AgentEnv example and setup guide that let developers run an evaluation using their existing Vercel CLI login, then inspect the task's score and the agent's recorded actions.

AgentEnv https://github.com/scaleapi/agentenv-framework , Scale AI's framework for testing agents against tasks and scoring their results, now runs on Vercel Sandbox https://vercel.com/docs/sandbox . The Vercel provider ships in AgentEnv 0.9.1334 and later, and runs the agent and its task environment in separate Sandboxes. Try the AgentEnv example https://github.com/vercel-labs/agentenv-framework/tree/vercel-demo to run an evaluation on Vercel Sandbox and inspect the task’s score and the agent’s recorded actions. After completing the setup guide https://vercel.com/docs/sandbox/ecosystem/agentenv , run this command from the cloned repository. It uses your existing Vercel CLI login: Replace YOUR TEAM with your Vercel team slug and YOUR PROJECT with an existing project name. Learn more in the Sandbox ecosystem documentation https://vercel.com/docs/sandbox/ecosystem .