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";
let calls = 0, token = "first";
vi.mock("./project-mcp-auth", () => ({ projectMcpAuthorization: async (server: object) => ({ server: { ...server, headers: { authorization: token } }, allowed: null }) }));
vi.mock("./project-mcp-transport", () => ({ withProjectMcpServer: async (_server: unknown, work: (rpc: () => Promise<unknown>) => Promise<unknown>) => {
calls++; return work(async () => ({ result: { tools: Array.from({ length: 151 }, (_,i) => ({ name: "tool_" + i, inputSchema: { type: "object" }, outputSchema: { type: "object" }, _meta: { ui: { resourceUri: "ui://fixture" } } })) } }));
} }));
import { listMcpToolPage } from "./project-mcp-catalog";
const server = { name: "fixture", transport: "http" as const, url: "https://example.com/mcp", headers: {}, oauthConfigured: false };
describe("complete bounded MCP discovery", () => {
it("paginates beyond 128 tools, retains schemas/UI metadata and reuses only matching auth cache", async () => {
const a = await listMcpToolPage(server, { limit: 100 });
expect(a.tools).toHaveLength(100); expect(a.complete).toBe(false);
const b = await listMcpToolPage(server, { cursor: a.nextCursor, limit: 100 });
expect(b.tools).toHaveLength(51); expect(b.complete).toBe(true); expect(calls).toBe(1);
expect(b.tools[0]).toMatchObject({ outputSchema: { type: "object" }, _meta: { ui: { resourceUri: "ui://fixture" } } });
token = "rotated";
await expect(listMcpToolPage(server, { cursor: a.nextCursor })).rejects.toThrow("another connection");
await listMcpToolPage(server); expect(calls).toBe(2);
});
});