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 { beforeEach, describe, expect, it, vi } from "vitest";
import type { WorkflowOptimizerCandidate } from "./graph-optimizer";
const executeIntegrationAction = vi.fn();
vi.mock("@/lib/infra/connection-dispatch", () => ({ executeIntegrationAction }));
const { createJevWorkflowOptimizerEvaluator } = await import("./jev-optimizer");
const candidate = (id: string): WorkflowOptimizerCandidate => ({
id,
kind: "presentation-group",
title: `Candidate ${id}`,
description: "Compact a reviewed workflow segment without changing authority.",
nodeIds: ["a", "b", "c"],
risk: "safe",
estimatedNodeDelta: 0,
hostEligible: true,
});
beforeEach(() => executeIntegrationAction.mockReset());
describe("Jev workflow optimizer adapter", () => {
it("adapts to state/questions typed evaluate schemas", async () => {
executeIntegrationAction
.mockResolvedValueOnce({ result: [{ name: "evaluate", inputSchema: { type: "object", properties: { state: {}, questions: {}, model: {} } } }] })
.mockResolvedValueOnce({ result: { content: [{ type: "text", text: JSON.stringify({ answers: { q1: { type: "noul", noul: 0.91 } } }) }] } });
const evaluate = createJevWorkflowOptimizerEvaluator({ user: "owner", connection: "jev", model: "jev-1.13.0" });
const result = await evaluate({ workflow: { nodeCount: 8 } }, [candidate("safe-1")]);
expect(result).toEqual({ provider: "jev", probabilities: { "safe-1": 0.91 } });
expect(executeIntegrationAction).toHaveBeenCalledTimes(2);
expect(executeIntegrationAction.mock.calls[1]?.[0]).toMatchObject({
provider: "mcp",
connection: "jev",
operation: "mcp.tool",
arguments: { name: "evaluate", arguments: { model: "jev-1.13.0" } },
});
});
it("adapts to jev_decide bounded-alternative schemas", async () => {
executeIntegrationAction
.mockResolvedValueOnce({ result: [{ name: "jev_decide", inputSchema: { type: "object", properties: { decision: {}, evidence: {}, priorities: {}, candidates: {}, requirements: {} } } }] })
.mockResolvedValueOnce({ result: { content: [{ type: "text", text: JSON.stringify({ recommendation: { selected: "apply", confidence: 0.88, probabilities: { apply: 0.84, skip: 0.16 } } }) }] } });
const evaluate = createJevWorkflowOptimizerEvaluator({ user: "owner", connection: "jev" });
const result = await evaluate({ workflow: { nodeCount: 8 } }, [candidate("safe-1")]);
expect(result.probabilities["safe-1"]).toBe(0.84);
expect(executeIntegrationAction.mock.calls[1]?.[0]).toMatchObject({
operation: "mcp.tool",
arguments: {
name: "jev_decide",
arguments: {
candidates: [
{ id: "apply", description: expect.any(String) },
{ id: "skip", description: expect.any(String) },
],
requirements: expect.any(Array),
},
},
});
});
it("fails closed on an unsupported advertised decision schema so the core can fallback", async () => {
executeIntegrationAction.mockResolvedValueOnce({
result: [{ name: "jev_decide", inputSchema: { type: "object", properties: { prompt: {} } } }],
});
const evaluate = createJevWorkflowOptimizerEvaluator({ user: "owner", connection: "jev" });
await expect(evaluate({}, [candidate("safe-1")])).rejects.toThrow("Unsupported Jev tool schema");
});
});