openclaw/extensions/codex/provider.test.ts

// Codex tests cover provider plugin behavior.
import { afterEach, describe, expect, it, vi } from "vitest";
import { CODEX_GPT5_BEHAVIOR_CONTRACT } from "./prompt-overlay.js";
import { buildCodexModelDefinition } from "./provider-catalog.js";
import { codexProviderDiscovery } from "./provider-discovery.js";
import {
  buildCodexProvider,
  buildCodexProviderCatalog,
  resolveCodexSupportedReasoningEffort,
} from "./provider.js";
import { CodexAppServerClient } from "./src/app-server/client.js";
import type { listCodexAppServerModels } from "./src/app-server/models.js";
import {
  createIsolatedCodexAppServerClient,
  getSharedCodexAppServerClient,
  resetSharedCodexAppServerClientForTests,
} from "./src/app-server/shared-client.js";

afterEach(() => {
  resetSharedCodexAppServerClientForTests();
  vi.restoreAllMocks();
});

function expectStaticFallbackCatalog(
  result: Awaited<ReturnType<typeof buildCodexProviderCatalog>>,
) {
  expect(result.provider.models.map((model) => model.id)).toEqual([
    "gpt-5.6-sol",
    "gpt-5.6-luna",
    "gpt-5.5",
    "gpt-5.4-mini",
  ]);
}

function createFakeCodexClient(): CodexAppServerClient {
  return {
    initialize: vi.fn(async () => undefined),
    request: vi.fn(async () => ({ data: [] })),
    addNotificationHandler: vi.fn(() => () => undefined),
    addRequestHandler: vi.fn(() => () => undefined),
    addCloseHandler: vi.fn(() => () => undefined),
    setThreadSessionRequestGuard: vi.fn(),
    close: vi.fn(),
  } as unknown as CodexAppServerClient;
}

const TEST_CODEX_APP_SERVER_CONFIG = {
  appServer: {
    command: "/tmp/openclaw-test-codex",
  },
};

async function listTestCodexAppServerModels(
  options: Parameters<typeof listCodexAppServerModels>[0] = {},
) {
  expect(options.sharedClient).toBe(false);
  const client = await createIsolatedCodexAppServerClient({
    startOptions: options.startOptions,
    timeoutMs: options.timeoutMs,
    authProfileId: null,
  });
  try {
    await client.request(
      "model/list",
      {
        limit: options.limit ?? null,
        cursor: options.cursor ?? null,
        includeHidden: options.includeHidden ?? null,
      },
      { timeoutMs: options.timeoutMs },
    );
    return { models: [] };
  } finally {
    client.close();
  }
}

function expectRecordFields(value: unknown, expected: Record<string, unknown>) {
  if (!value || typeof value !== "object") {
    throw new Error("Expected record");
  }
  const actual = value as Record<string, unknown>;
  for (const [key, expectedValue] of Object.entries(expected)) {
    expect(actual[key]).toEqual(expectedValue);
  }
  return actual;
}

function mockCallArg(mockFn: { mock: { calls: unknown[][] } }, callIndex: number): unknown {
  return mockFn.mock.calls[callIndex]?.[0];
}

describe("codex provider", () => {
  it.each(["gpt-5.6-sol", "gpt-5.6-terra", "gpt-5.6-luna"])(
    "uses the known context window for discovered %s models",
    (modelId) => {
      const model = buildCodexModelDefinition({
        id: modelId,
        model: modelId,
        inputModalities: ["text", "image"],
      });

      expect(model.contextWindow).toBe(372_000);
    },
  );

  it.each(["gpt-5.5-pro", "gpt-5.4-pro"] as const)(
    "classifies %s as a modern Codex model",
    (modelId) => {
      const provider = buildCodexProvider();

      expect(
        provider.isModernModelRef?.({
          provider: "openai",
          modelId,
        } as never),
      ).toBe(true);
    },
  );

  it("maps Codex app-server models to a Codex provider catalog", async () => {
    const listModels = vi.fn(async () => ({
      models: [
        {
          id: "gpt-5.4",
          model: "gpt-5.4",
          displayName: "gpt-5.4",
          hidden: false,
          inputModalities: ["text", "image"],
          supportedReasoningEfforts: ["low", "medium", "high", "xhigh"],
        },
        {
          id: "hidden-model",
          model: "hidden-model",
          hidden: true,
          inputModalities: ["text"],
          supportedReasoningEfforts: [],
        },
      ],
    }));

    const result = await buildCodexProviderCatalog({
      env: {},
      listModels,
      pluginConfig: { discovery: { timeoutMs: 1234 } },
    });

    expectRecordFields(mockCallArg(listModels, 0), {
      limit: 100,
      timeoutMs: 1234,
      sharedClient: false,
    });
    expectRecordFields(result.provider, {
      auth: "token",
      api: "openai-chatgpt-responses",
    });
    expect(result.provider.models).toHaveLength(1);
    expectRecordFields(result.provider.models[0], {
      id: "gpt-5.4",
      name: "gpt-5.4",
      reasoning: true,
      input: ["text", "image"],
      compat: {
        supportsReasoningEffort: true,
        supportedReasoningEfforts: ["low", "medium", "high", "xhigh"],
        supportsUsageInStreaming: true,
      },
    });
  });

  it("keeps a static fallback catalog when discovery is disabled", async () => {
    const listModels = vi.fn();

    const result = await buildCodexProviderCatalog({
      env: {},
      listModels,
      pluginConfig: { discovery: { enabled: false } },
    });

    expect(listModels).not.toHaveBeenCalled();
    expectStaticFallbackCatalog(result);
  });

  it("uses live plugin config to re-enable discovery after startup disable", async () => {
    const listModels = vi.fn(async () => ({
      models: [
        {
          id: "gpt-5.4",
          model: "gpt-5.4",
          displayName: "gpt-5.4",
          hidden: false,
          inputModalities: ["text", "image"],
          supportedReasoningEfforts: ["low", "medium", "high", "xhigh"],
        },
      ],
    }));
    const provider = buildCodexProvider({
      pluginConfig: { discovery: { enabled: false } },
      listModels,
    });

    const result = await provider.catalog?.run({
      config: {
        plugins: {
          entries: {
            codex: {
              config: {
                discovery: {
                  enabled: true,
                  timeoutMs: 4321,
                },
              },
            },
          },
        },
      },
      env: {},
    } as never);

    expectRecordFields(mockCallArg(listModels, 0), {
      limit: 100,
      timeoutMs: 4321,
      sharedClient: false,
    });
    const resultProvider = result && "provider" in result ? result.provider : undefined;
    expect(resultProvider?.models.map((model) => model.id)).toEqual(["gpt-5.4"]);
  });

  it("pages through live discovery before building the provider catalog", async () => {
    const listModels = vi
      .fn()
      .mockResolvedValueOnce({
        models: [
          {
            id: "gpt-5.4",
            model: "gpt-5.4",
            hidden: false,
            inputModalities: ["text", "image"],
            supportedReasoningEfforts: ["medium"],
          },
        ],
        nextCursor: "page-2",
      })
      .mockResolvedValueOnce({
        models: [
          {
            id: "gpt-5.5",
            model: "gpt-5.5",
            hidden: false,
            inputModalities: ["text"],
            supportedReasoningEfforts: [],
          },
        ],
      });

    const result = await buildCodexProviderCatalog({
      env: {},
      listModels,
    });

    expectRecordFields(mockCallArg(listModels, 0), {
      cursor: undefined,
      limit: 100,
      sharedClient: false,
    });
    expectRecordFields(mockCallArg(listModels, 1), {
      cursor: "page-2",
      limit: 100,
      sharedClient: false,
    });
    expect(result.provider.models.map((model) => model.id)).toEqual(["gpt-5.4", "gpt-5.5"]);
  });

  it("reports discovery failures before using the fallback catalog", async () => {
    const error = new Error("app-server down");
    const onDiscoveryFailure = vi.fn();
    const listModels = vi.fn(async () => {
      throw error;
    });

    const result = await buildCodexProviderCatalog({
      env: {},
      listModels,
      onDiscoveryFailure,
    });

    expect(onDiscoveryFailure).toHaveBeenCalledWith(error);
    expectStaticFallbackCatalog(result);
  });

  it("keeps a static fallback catalog when live discovery is explicitly disabled by env", async () => {
    const listModels = vi.fn();

    const result = await buildCodexProviderCatalog({
      env: { OPENCLAW_CODEX_DISCOVERY_LIVE: "0" },
      listModels,
    });

    expect(listModels).not.toHaveBeenCalled();
    expectStaticFallbackCatalog(result);
  });

  it("closes the transient app-server client after live discovery", async () => {
    const client = createFakeCodexClient();
    vi.spyOn(CodexAppServerClient, "start").mockReturnValue(client);

    await buildCodexProviderCatalog({
      env: { OPENCLAW_CODEX_DISCOVERY_LIVE: "1" },
      pluginConfig: TEST_CODEX_APP_SERVER_CONFIG,
      listModels: listTestCodexAppServerModels,
    });

    expect(client["close"]).toHaveBeenCalledTimes(1);
  });

  it("does not close an active shared app-server client during live discovery", async () => {
    const activeClient = createFakeCodexClient();
    const discoveryClient = createFakeCodexClient();
    vi.spyOn(CodexAppServerClient, "start")
      .mockReturnValueOnce(activeClient)
      .mockReturnValueOnce(discoveryClient);

    await getSharedCodexAppServerClient({
      startOptions: {
        transport: "stdio",
        command: "/tmp/openclaw-test-codex",
        commandSource: "config",
        args: ["app-server", "--listen", "stdio://"],
        headers: {},
      },
      timeoutMs: 1000,
      authProfileId: null,
    });
    await buildCodexProviderCatalog({
      env: { OPENCLAW_CODEX_DISCOVERY_LIVE: "1" },
      pluginConfig: TEST_CODEX_APP_SERVER_CONFIG,
      listModels: listTestCodexAppServerModels,
    });

    expect(activeClient["close"]).not.toHaveBeenCalled();
    expect(discoveryClient["close"]).toHaveBeenCalledTimes(1);
  });

  it("resolves arbitrary Codex app-server model ids as text-only until discovered", () => {
    const provider = buildCodexProvider();

    const model = provider.resolveDynamicModel?.({
      provider: "codex",
      modelId: " custom-model ",
      modelRegistry: { find: () => null },
    } as never);

    expectRecordFields(model, {
      id: "custom-model",
      provider: "codex",
      api: "openai-chatgpt-responses",
      baseUrl: "https://chatgpt.com/backend-api",
      input: ["text"],
    });
  });

  it("keeps fallback Codex app-server models image-capable", () => {
    const provider = buildCodexProvider();

    const model = provider.resolveDynamicModel?.({
      provider: "codex",
      modelId: "gpt-5.5",
      modelRegistry: { find: () => null },
    } as never);

    expectRecordFields(model, {
      id: "gpt-5.5",
      input: ["text", "image"],
    });
  });

  it("treats o4 ids as reasoning-capable Codex models", () => {
    const provider = buildCodexProvider();

    const model = provider.resolveDynamicModel?.({
      provider: "codex",
      modelId: "o4-mini",
      modelRegistry: { find: () => null },
    } as never);

    expectRecordFields(model, {
      id: "o4-mini",
      reasoning: true,
      compat: {
        supportsUsageInStreaming: true,
      },
    });
    expect(
      provider
        .resolveThinkingProfile?.({ provider: "codex", modelId: "o4-mini" } as never)
        ?.levels.some((level) => level.id === "xhigh"),
    ).toBe(true);
  });

  it("uses fallback reasoning metadata for GPT-5.6 Luna", () => {
    const provider = buildCodexProvider();
    const model = provider.resolveDynamicModel?.({
      provider: "codex",
      modelId: "gpt-5.6-luna",
      modelRegistry: { find: () => null },
    } as never);

    expectRecordFields(model, {
      id: "gpt-5.6-luna",
      reasoning: true,
      contextWindow: 372_000,
      compat: {
        supportsReasoningEffort: true,
        supportedReasoningEfforts: ["low", "medium", "high", "xhigh", "max"],
        supportsUsageInStreaming: true,
      },
    });
    expect(
      provider
        .resolveThinkingProfile?.({ provider: "codex", modelId: "gpt-5.6-luna" } as never)
        ?.levels.map((level) => level.id),
    ).toContain("max");
  });

  it("exposes max only for known native GPT-5.6 models", () => {
    const provider = buildCodexProvider();
    const levels = (modelId: string) =>
      provider
        .resolveThinkingProfile?.({ provider: "codex", modelId } as never)
        ?.levels.map((level) => level.id);

    expect(levels("gpt-5.6-sol")).toContain("max");
    expect(levels("gpt-5.6-terra")).toContain("max");
    expect(levels("gpt-5.6-luna")).toContain("max");
    expect(levels("gpt-5.6")).not.toContain("max");
    expect(levels("gpt-5.6-sol-oai")).not.toContain("max");
  });

  it("uses app-server reasoning metadata as the authoritative thinking profile", () => {
    const provider = buildCodexProvider();

    expect(
      provider
        .resolveThinkingProfile?.({
          provider: "codex",
          modelId: "gpt-5.4-pro",
          compat: { supportedReasoningEfforts: ["medium", "high", "xhigh"] },
        } as never)
        ?.levels.map((level) => level.id),
    ).toEqual(["off", "medium", "high", "xhigh"]);
  });

  it("uses known GPT-5.6 native Codex fallbacks when model/list metadata is unavailable", () => {
    const provider = buildCodexProvider();
    const levels = (modelId: string, supportedReasoningEfforts?: string[]) =>
      provider
        .resolveThinkingProfile?.({
          provider: "codex",
          modelId,
          ...(supportedReasoningEfforts ? { compat: { supportedReasoningEfforts } } : {}),
        } as never)
        ?.levels.map((level) => level.id);

    expect(levels("gpt-5.6-sol")).toContain("ultra");
    expect(levels("gpt-5.6-terra")).toContain("ultra");
    expect(levels("gpt-5.6-luna")).toEqual(["off", "low", "medium", "high", "xhigh", "max"]);
    expect(levels("gpt-5.6")).not.toContain("ultra");

    const directOpenAIEfforts = ["none", "low", "medium", "high", "xhigh", "max"];
    expect(levels("gpt-5.6-sol", directOpenAIEfforts)).toContain("ultra");
    expect(levels("gpt-5.6-terra", directOpenAIEfforts)).toContain("ultra");
  });

  it.each([
    { modelId: "gpt-5.6-sol", expected: "low" },
    { modelId: "gpt-5.6-terra", expected: "medium" },
    { modelId: "gpt-5.6-luna", expected: "medium" },
  ] as const)("uses the native $modelId default reasoning effort", ({ modelId, expected }) => {
    const provider = buildCodexProvider();

    expect(
      provider.resolveThinkingProfile?.({ provider: "codex", modelId } as never)?.defaultLevel,
    ).toBe(expected);
  });

  it("omits the native default when authoritative model/list metadata does not support it", () => {
    const provider = buildCodexProvider();

    expect(
      provider.resolveThinkingProfile?.({
        provider: "codex",
        modelId: "gpt-5.6-sol",
        compat: { supportedReasoningEfforts: ["high"] },
      } as never)?.defaultLevel,
    ).toBeUndefined();
  });

  it("uses app-server model/list reasoning metadata as authoritative", () => {
    const provider = buildCodexProvider();
    const levels = (modelId: string, supportedReasoningEfforts: string[]) =>
      provider
        .resolveThinkingProfile?.({
          provider: "codex",
          modelId,
          compat: { supportedReasoningEfforts },
        } as never)
        ?.levels.map((level) => level.id);

    const maxEfforts = ["low", "medium", "high", "xhigh", "max"];
    const ultraEfforts = [...maxEfforts, "ultra"];
    expect(levels("gpt-5.6-sol", maxEfforts)).not.toContain("ultra");
    expect(levels("gpt-5.6-terra", maxEfforts)).not.toContain("ultra");
    expect(levels("gpt-5.6-sol", ultraEfforts)).toContain("ultra");
    expect(levels("gpt-5.6-terra", ultraEfforts)).toContain("ultra");
    expect(levels("gpt-5.6-luna", maxEfforts)).not.toContain("ultra");
  });

  it.each([
    ["max", ["low", "medium", "high", "xhigh", "ultra"], "xhigh"],
    ["xhigh", ["low", "medium", "high", "ultra"], "high"],
  ] as const)(
    "does not upgrade requested %s to Ultra when model metadata omits that effort",
    (requested, supportedReasoningEfforts, expected) => {
      expect(resolveCodexSupportedReasoningEffort({ requested, supportedReasoningEfforts })).toBe(
        expected,
      );
    },
  );

  it.each(["gpt-5.5-pro", "gpt-5.4-pro"] as const)(
    "uses the known %s effort profile when app-server metadata is absent",
    (modelId) => {
      const provider = buildCodexProvider();

      expect(
        provider
          .resolveThinkingProfile?.({
            provider: "codex",
            modelId,
          } as never)
          ?.levels.map((level) => level.id),
      ).toEqual(["off", "medium", "high", "xhigh"]);
    },
  );

  it("declares synthetic auth because the harness owns Codex credentials", () => {
    const provider = buildCodexProvider();

    expect(provider.resolveSyntheticAuth?.({ provider: "codex" })).toEqual({
      apiKey: "codex-app-server",
      source: "codex-app-server",
      mode: "token",
    });
  });

  // Mirrors CODEX_APP_SERVER_AUTH_MARKER without a scanner-visible token literal.
  function appServerMarkerToken(): string {
    return ["codex", "app", "server"].join("-");
  }

  it("fetches usage from native Codex app-server rate limits for synthetic auth", async () => {
    const readUsage = vi.fn(async () => ({
      rateLimits: {
        rateLimitsByLimitId: {
          codex: {
            limitId: "codex",
            primary: {
              usedPercent: 9,
              windowDurationMins: 300,
              resetsAt: 1_700_003_600,
            },
          },
        },
      },
      accountEmail: "codex-account@example.com",
    }));
    const provider = buildCodexProvider({ readUsage });

    await expect(
      provider.fetchUsageSnapshot?.({
        provider: "openai",
        token: "codex-app-server",
        timeoutMs: 3500,
        config: {},
        env: {},
        fetchFn: fetch,
      } as never),
    ).resolves.toEqual({
      provider: "openai",
      displayName: "OpenAI",
      windows: [{ label: "5h", usedPercent: 9, resetAt: 1_700_003_600_000 }],
      plan: undefined,
      accountEmail: "codex-account@example.com",
    });
    expect(readUsage).toHaveBeenCalledWith({
      timeoutMs: 3500,
      agentDir: undefined,
      config: {},
      startOptions: expect.objectContaining({
        command: "codex",
        commandSource: "managed",
      }),
    });
  });

  it("keeps synthetic usage reads on the configured Codex auth bridge", async () => {
    const scopedRequest = vi.fn(async ({ method }: { method: string }) =>
      method === "account/rateLimits/read"
        ? { rateLimitsByLimitId: {} }
        : { account: { email: "bridge@example.com" } },
    );
    const withCodexAppServerJsonClient = vi.fn(
      async (
        _params: unknown,
        run: (request: typeof scopedRequest) => Promise<unknown>,
      ): Promise<unknown> => await run(scopedRequest),
    );
    vi.doMock("./src/app-server/request.js", () => ({
      withCodexAppServerJsonClient,
    }));
    try {
      const provider = buildCodexProvider();

      await provider.fetchUsageSnapshot?.({
        provider: "openai",
        token: "codex-app-server",
        authProfileId: "openai:work",
        timeoutMs: 3500,
        config: {
          plugins: {
            entries: {
              codex: {
                config: TEST_CODEX_APP_SERVER_CONFIG,
              },
            },
          },
        },
        env: {},
        fetchFn: fetch,
      } as never);

      expect(withCodexAppServerJsonClient).toHaveBeenCalledWith(
        expect.objectContaining({
          timeoutMs: 3500,
          authProfileId: "openai:work",
          config: {
            plugins: {
              entries: {
                codex: {
                  config: TEST_CODEX_APP_SERVER_CONFIG,
                },
              },
            },
          },
          startOptions: expect.objectContaining({
            command: "/tmp/openclaw-test-codex",
            commandSource: "config",
            args: ["app-server", "--listen", "stdio://"],
          }),
          isolated: true,
          isolatedShutdown: { forceKillDelayMs: 200, exitTimeoutMs: 300 },
        }),
        expect.any(Function),
      );
      expect(scopedRequest).toHaveBeenCalledWith({ method: "account/rateLimits/read" });
      expect(scopedRequest).toHaveBeenCalledWith({ method: "account/read", requestParams: {} });
    } finally {
      vi.doUnmock("./src/app-server/request.js");
    }
  });

  it("keeps the rate-limit windows when the account identity read fails", async () => {
    const readUsage = vi.fn(async () => ({
      rateLimits: {
        rateLimitsByLimitId: {
          codex: {
            limitId: "codex",
            primary: { usedPercent: 12, windowDurationMins: 300 },
          },
        },
      },
    }));
    const provider = buildCodexProvider({ readUsage });

    await expect(
      provider.fetchUsageSnapshot?.({
        provider: "openai",
        token: appServerMarkerToken(),
        timeoutMs: 3500,
        config: {},
        env: {},
        fetchFn: fetch,
      } as never),
    ).resolves.toEqual({
      provider: "openai",
      displayName: "OpenAI",
      windows: [{ label: "5h", usedPercent: 12 }],
      plan: undefined,
    });
  });

  it("returns the rate-limit windows when the account identity read hangs", async () => {
    const scopedRequest = vi.fn(({ method }: { method: string }) =>
      method === "account/rateLimits/read"
        ? Promise.resolve({
            rateLimitsByLimitId: {
              codex: {
                limitId: "codex",
                primary: { usedPercent: 7, windowDurationMins: 300 },
              },
            },
          })
        : // account/read never settles; the best-effort bound must win.
          new Promise<never>(() => {}),
    );
    const withCodexAppServerJsonClient = vi.fn(
      async (
        _params: unknown,
        run: (request: typeof scopedRequest) => Promise<unknown>,
      ): Promise<unknown> => await run(scopedRequest),
    );
    vi.doMock("./src/app-server/request.js", () => ({
      withCodexAppServerJsonClient,
    }));
    try {
      const provider = buildCodexProvider();
      // A tiny usage budget drives the identity bound to a few real
      // milliseconds so the hung account read is dropped promptly.
      await expect(
        provider.fetchUsageSnapshot?.({
          provider: "openai",
          token: appServerMarkerToken(),
          timeoutMs: 900,
          config: {},
          env: {},
          fetchFn: fetch,
        } as never),
      ).resolves.toEqual({
        provider: "openai",
        displayName: "OpenAI",
        windows: [{ label: "5h", usedPercent: 7 }],
        plan: undefined,
      });
    } finally {
      vi.doUnmock("./src/app-server/request.js");
    }
  });

  it("exposes a setup auth choice for installing Codex as an external provider", async () => {
    const provider = buildCodexProvider();

    const authChoice = provider.auth[0];
    expectRecordFields(authChoice, {
      id: "app-server",
      kind: "custom",
    });
    expectRecordFields(authChoice?.wizard, {
      choiceId: "codex",
      choiceLabel: "Codex app-server",
      onboardingScopes: ["text-inference"],
    });
    const authResult = await authChoice?.run({} as never);
    expectRecordFields(authResult, {
      profiles: [],
      defaultModel: "codex/gpt-5.6-sol",
    });
  });

  it("exposes a lightweight provider-discovery entry for model list/status", async () => {
    expect(codexProviderDiscovery.id).toBe("codex");
    expect(codexProviderDiscovery.resolveSyntheticAuth?.({ provider: "codex" })).toEqual({
      apiKey: "codex-app-server",
      source: "codex-app-server",
      mode: "token",
    });

    const result = await codexProviderDiscovery.staticCatalog?.run({
      config: {},
      env: {},
      agentDir: "/tmp/openclaw-agent",
    } as never);

    const models = result && "provider" in result ? result.provider.models : [];
    expect(models.map((model) => model.id)).toEqual([
      "gpt-5.6-sol",
      "gpt-5.6-luna",
      "gpt-5.5",
      "gpt-5.4-mini",
    ]);
    expect(models.find((model) => model.id === "gpt-5.6-sol")?.contextWindow).toBe(372_000);
    expect(models.find((model) => model.id === "gpt-5.6-luna")?.contextWindow).toBe(372_000);
  });

  it("adds the GPT-5 prompt overlay to Codex provider runs", () => {
    const provider = buildCodexProvider();

    const contribution = provider.resolveSystemPromptContribution?.({
      provider: "codex",
      modelId: "gpt-5.4",
    } as never);
    expectRecordFields(contribution, {
      stablePrefix: CODEX_GPT5_BEHAVIOR_CONTRACT,
    });
    const interactionStyle = contribution?.sectionOverrides?.interaction_style;
    expect(interactionStyle).toContain("Live chat: short, natural, human.");
    expect(interactionStyle).not.toContain("Heartbeat = useful proactive progress");
  });

  it("does not add the GPT-5 prompt overlay to non-GPT-5 Codex provider runs", () => {
    const provider = buildCodexProvider();

    expect(
      provider.resolveSystemPromptContribution?.({
        provider: "codex",
        modelId: "o4-mini",
      } as never),
    ).toBeUndefined();
  });
});