Files
oh-my-pi/packages/coding-agent/test/sdk-computer-tool-toggle.test.ts
T
can1357 681d7daf65 feat(computer): unified native addon, /computer toggle, function tool
- Replaced the separate GUI-linked pi_natives.desktop.linux-x64 addon with
  a pure-Rust X11 backend (x11rb RustConnection capture via RandR/GetImage,
  XTest input with keysym mapping) compiled into the core addon on every
  published target; Linux arm64 and musl are now supported and headless
  hosts load the addon unaffected.
- Removed the native-desktop-linux cargo feature, desktop_unsupported.rs,
  lazy desktop loader, second napi build, desktop packaging/CI steps, GUI
  build dependencies, and the now-unreferenced vendored libspa crate;
  reverted setup-system-deps to main.
- Preserved the desktop input hardening semantics on the unified backend:
  XTest layouts reject negative origins and coordinates beyond 0..=32767,
  batch coordinates stay bound to the frame last returned to JS with
  intermediate screenshots deferred, coordinate input requires a
  previously returned frame, and failed chord releases still release
  every held key.
- Enforced a 60s worker-side execute deadline (DESKTOP_DEADLINE_EXCEEDED):
  no input is emitted after expiry and wait-heavy batches are rejected
  upfront.
- Added int32 fail-closed validation for coordinates, drag points, and
  scroll deltas at the JS ingress and gateway schema.
- Exposed computer to models without native OpenAI computer-use support as
  a regular function tool with a typed GA action schema across OpenAI,
  Azure, and Codex Responses providers, including named forced choice.
- Added the /computer slash command (on/off/status/toggle) for
  session-only enablement via runtime tool registration in SessionTools.
- Updated docs, changelogs, and contract tests accordingly.
2026-07-24 01:40:05 +00:00

76 lines
3.4 KiB
TypeScript

import { afterAll, beforeAll, describe, expect, it } from "bun:test";
import * as fs from "node:fs";
import * as os from "node:os";
import * as path from "node:path";
import { AuthStorage } from "@oh-my-pi/pi-ai";
import { getBundledModel } from "@oh-my-pi/pi-catalog/models";
import { ModelRegistry } from "@oh-my-pi/pi-coding-agent/config/model-registry";
import { Settings } from "@oh-my-pi/pi-coding-agent/config/settings";
import { createAgentSession } from "@oh-my-pi/pi-coding-agent/sdk";
import type { AgentSession } from "@oh-my-pi/pi-coding-agent/session/agent-session";
import { SessionManager } from "@oh-my-pi/pi-coding-agent/session/session-manager";
import { removeSyncWithRetries, Snowflake } from "@oh-my-pi/pi-utils";
// Guards the /computer session-scoped toggle mechanism: createTools derives the
// built-in slate once at session start, so with `computer.enabled=false` the
// computer tool is entirely absent from the registry. `setComputerToolEnabled`
// must re-derive that one entry through the SDK-provided factory and flip the
// active slate for the next turn — without persisting anything. Constructing
// the tool is inert (the desktop worker only spawns on first execute), so this
// runs headless.
describe("AgentSession.setComputerToolEnabled", () => {
let registryDir: string;
let authStorage: AuthStorage;
let modelRegistry: ModelRegistry;
const sessions: AgentSession[] = [];
beforeAll(async () => {
registryDir = path.join(os.tmpdir(), `pi-computer-toggle-${Snowflake.next()}`);
fs.mkdirSync(registryDir, { recursive: true });
authStorage = await AuthStorage.create(path.join(registryDir, "auth.db"));
modelRegistry = new ModelRegistry(authStorage);
});
afterAll(async () => {
for (const session of sessions) await session.dispose().catch(() => {});
authStorage.close();
if (fs.existsSync(registryDir)) removeSyncWithRetries(registryDir);
});
it("flips the active tool slate on enable and disable, session-only", async () => {
const settings = Settings.isolated({});
const { session } = await createAgentSession({
cwd: registryDir,
agentDir: registryDir,
modelRegistry,
sessionManager: SessionManager.inMemory(),
settings,
model: getBundledModel("openai", "gpt-4o-mini"),
disableExtensionDiscovery: true,
});
sessions.push(session);
// computer.enabled defaults to false: absent from registry and slate.
expect(session.getAllToolNames()).not.toContain("computer");
expect(session.getActiveToolNames()).not.toContain("computer");
// The /computer command's enable path: slate refresh + runtime override.
expect(await session.setComputerToolEnabled(true)).toBe(true);
session.settings.override("computer.enabled", true);
expect(session.getAllToolNames()).toContain("computer");
expect(session.getEnabledToolNames()).toContain("computer");
expect(session.settings.get("computer.enabled")).toBe(true);
// Disable removes it from the active slate but keeps the registry entry,
// so a re-enable never registers a second desktop controller.
expect(await session.setComputerToolEnabled(false)).toBe(true);
session.settings.override("computer.enabled", false);
expect(session.getEnabledToolNames()).not.toContain("computer");
expect(session.getAllToolNames()).toContain("computer");
// Re-enable reuses the retained registry entry.
expect(await session.setComputerToolEnabled(true)).toBe(true);
expect(session.getEnabledToolNames()).toContain("computer");
});
});