Browser-based graphical shell and control plane for a Linux server you own — real terminal, file manager, live metrics and BYOK AI in one mobile-first workspace. Next.js 16 · React 19 · node-pty · Tailwind v4 · shadcn/ui. Public alpha, self-hosted.
TypeScript browser-shellcontrol-planefile-managerhomelabmobile-firstnextjsptyreactself-hostedserver-managementshadcn-uisysadmintailwindcssterminaltypescriptvpsweb-desktopweb-terminal
import { describe, expect, it, vi } from "vitest";
vi.mock("server-only", () => ({}));
const { aggregateTaskQuality, externalAgentContract, runTaskQualityFixture } = await import("./bench-task-quality.mjs");
describe("task quality benchmark", () => {
it("scores the deterministic telemetry fixture exactly", () => {
expect(runTaskQualityFixture()).toMatchObject({ passed: true, deterministic: true, quality: { retries: 2, invalidArgSteps: 1, rollbackSignals: 1, averageStepDurationMs: 25 } });
});
it("aggregates recipe outcomes and telemetry without inventing coverage for legacy recipes", () => {
const result = aggregateTaskQuality([
{ attempts: 3, successes: 2, failures: 1, qualityVersion: 1, quality: { stepAttempts: 10, completedSteps: 8, failedSteps: 1, deniedSteps: 0, rateLimitedSteps: 0, invalidArgSteps: 1, retries: 2, rollbackSignals: 1, timedSteps: 4, totalStepDurationMs: 80, averageStepDurationMs: 20 } },
{ attempts: 1, successes: 1, failures: 0 },
]);
expect(result).toMatchObject({ recipes: 2, telemetryRecipes: 1, telemetryCoveragePct: 50, workflowSuccessPct: 75, toolCompletionPct: 80, invalidArgPct: 10, retryPct: 20, averageStepDurationMs: 20 });
});
it("requires the same model and exposes non-interactive runners for all agents", () => {
expect(externalAgentContract.requiresSameModel).toBe(true);
expect(externalAgentContract.runners.mso.comparable).toBe(true);
expect(externalAgentContract.runners.hermes.comparable).toBe(true);
expect(externalAgentContract.runners.openclaw.comparable).toBe(true);
});
});