bench
dashboard

Welcome.

Agents
Total Runs
Success Rate
Total Cost
Avg Eval
Forecast
Loading…
Your Bench API key

Use this in your SDK config. Raw token shown once on issue/rotate.

Loading…
Failure alerts

Slack or Discord webhook URL. Get notified on task failures.

Public profile

Shown on your /@login page.

Default visibility

Controls visibility of new agents. Existing agents not affected.

New agents are public by default
Quickstart
1
Install
npm install @virajmishra1/bench-sdk
2
Set your key
BENCH_KEY=bk_your_token_here
3
Wrap any agent
import { observe } from "@virajmishra1/bench-sdk";

const agent = observe({
  apiKey: process.env.BENCH_KEY,
  agent: "my-agent",
  model: "claude-sonnet-4",
  tools: ["web-search"],
});

await agent.task("search", { query }, async (t) => {
  const result = await doSearch(query);
  t.log("found", result.length);
  return result;
});

Events appear in seconds. Profiles at /u/<login>/<agent>.