QA Lab: add provider registry

Move mock and live provider behavior behind provider-owned definitions so suite, manual, Matrix, and transport lanes share defaults, auth staging, model config, and standalone server startup. Add AIMock as a first-class local provider mode while keeping mock-openai as the scenario-aware deterministic lane.
2026-05-06 11:20:43 +00:00 · 2026-04-16 21:23:23 -04:00
parent 4acab55db8
commit bb7e9823a8
60 changed files with 1964 additions and 998 deletions
--- a/docs/concepts/qa-e2e-automation.md
+++ b/docs/concepts/qa-e2e-automation.md
@@ -151,6 +151,22 @@ The baseline list should stay broad enough to cover:
 - repo-reading and docs-reading
 - one small build task such as Lobster Invaders

+## Provider mock lanes
+
+`qa suite` has two local provider mock lanes:
+
+- `mock-openai` is the scenario-aware OpenClaw mock. It remains the default
+  deterministic mock lane for repo-backed QA and parity gates.
+- `aimock` starts an AIMock-backed provider server for experimental protocol,
+  fixture, record/replay, and chaos coverage. It is additive and does not
+  replace the `mock-openai` scenario dispatcher.
+
+Provider-lane implementation lives under `extensions/qa-lab/src/providers/`.
+Each provider owns its defaults, local server startup, gateway model config,
+auth-profile staging needs, and live/mock capability flags. Shared suite and
+gateway code should route through the provider registry instead of branching on
+provider names.
+
 ## Transport adapters

 `qa-lab` owns a generic transport seam for markdown QA scenarios.
--- a/docs/help/testing.md
+++ b/docs/help/testing.md
@@ -52,6 +52,10 @@ These commands sit beside the main test suites when you need QA-lab realism:
    gateway workers, up to 64 workers or the selected scenario count. Use
    `--concurrency <count>` to tune the worker count, or `--concurrency 1` for
    the older serial lane.
+  - Supports provider modes `live-frontier`, `mock-openai`, and `aimock`.
+    `aimock` starts a local AIMock-backed provider server for experimental
+    fixture and protocol-mock coverage without replacing the scenario-aware
+    `mock-openai` lane.
 - `pnpm openclaw qa suite --runner multipass`
  - Runs the same QA suite inside a disposable Multipass Linux VM.
  - Keeps the same scenario-selection behavior as `qa suite` on the host.
@@ -65,6 +69,9 @@ These commands sit beside the main test suites when you need QA-lab realism:
    `.artifacts/qa-e2e/...`.
 - `pnpm qa:lab:up`
  - Starts the Docker-backed QA site for operator-style QA work.
+- `pnpm openclaw qa aimock`
+  - Starts only the local AIMock provider server for direct protocol smoke
+    testing.
 - `pnpm openclaw qa matrix`
  - Runs the Matrix live QA lane against a disposable Docker-backed Tuwunel homeserver.
  - This QA host is repo/dev-only today. Packaged OpenClaw installs do not ship
--- a/extensions/qa-lab/api.ts
+++ b/extensions/qa-lab/api.ts
@@ -6,7 +6,6 @@ export * from "./src/cli.js";
 export * from "./src/harness-runtime.js";
 export * from "./src/lab-server.js";
 export * from "./src/docker-harness.js";
-export * from "./src/mock-openai-server.js";
 export * from "./src/qa-agent-bootstrap.js";
 export * from "./src/qa-agent-workspace.js";
 export * from "./src/qa-gateway-config.js";
--- a/extensions/qa-lab/src/character-eval.ts
+++ b/extensions/qa-lab/src/character-eval.ts
@@ -3,38 +3,28 @@ import path from "node:path";
 import { formatErrorMessage } from "openclaw/plugin-sdk/error-runtime";
 import { runQaManualLane } from "./manual-lane.runtime.js";
 import { isQaFastModeModelRef, type QaProviderMode } from "./model-selection.js";
+import {
+  QA_FRONTIER_CHARACTER_EVAL_MODELS,
+  QA_FRONTIER_CHARACTER_JUDGE_MODEL_OPTIONS,
+  QA_FRONTIER_CHARACTER_JUDGE_MODELS,
+  QA_FRONTIER_CHARACTER_THINKING_BY_MODEL,
+} from "./providers/live-frontier/character-eval.js";
 import { type QaThinkingLevel } from "./qa-gateway-config.js";
 import { extractQaVisibleReplyLeakText } from "./reply-failure.js";
 import { runQaSuiteFromRuntime } from "./suite-launch.runtime.js";
 import type { QaSuiteResult } from "./suite.js";

 const DEFAULT_CHARACTER_SCENARIO_ID = "character-vibes-gollum";
-const DEFAULT_CHARACTER_EVAL_MODELS = Object.freeze([
-  "openai/gpt-5.4",
-  "openai/gpt-5.2",
-  "openai/gpt-5",
-  "anthropic/claude-opus-4-6",
-  "anthropic/claude-sonnet-4-6",
-  "zai/glm-5.1",
-  "moonshot/kimi-k2.5",
-  "google/gemini-3.1-pro-preview",
-]);
+const DEFAULT_CHARACTER_EVAL_MODELS = QA_FRONTIER_CHARACTER_EVAL_MODELS;
 const DEFAULT_CHARACTER_THINKING: QaThinkingLevel = "high";
 const DEFAULT_CHARACTER_EVAL_CONCURRENCY = 16;
 const DEFAULT_CHARACTER_THINKING_BY_MODEL: Readonly<Record<string, QaThinkingLevel>> =
-  Object.freeze({
-    "openai/gpt-5.4": "xhigh",
-    "openai/gpt-5.2": "xhigh",
-    "openai/gpt-5": "xhigh",
-  });
-const DEFAULT_JUDGE_MODELS = Object.freeze(["openai/gpt-5.4", "anthropic/claude-opus-4-6"]);
+  QA_FRONTIER_CHARACTER_THINKING_BY_MODEL;
+const DEFAULT_JUDGE_MODELS = QA_FRONTIER_CHARACTER_JUDGE_MODELS;
 const DEFAULT_JUDGE_THINKING: QaThinkingLevel = "xhigh";
 const DEFAULT_JUDGE_TIMEOUT_MS = 300_000;
 const DEFAULT_JUDGE_MODEL_OPTIONS: Readonly<Record<string, QaCharacterModelOptions>> =
-  Object.freeze({
-    "openai/gpt-5.4": { thinkingDefault: "xhigh" },
-    "anthropic/claude-opus-4-6": { thinkingDefault: "high" },
-  });
+  QA_FRONTIER_CHARACTER_JUDGE_MODEL_OPTIONS;

 type QaCharacterRunStatus = "pass" | "fail";

--- a/extensions/qa-lab/src/cli.runtime.test.ts
+++ b/extensions/qa-lab/src/cli.runtime.test.ts
@@ -24,14 +24,8 @@ const {
  writeQaDockerHarnessFiles: vi.fn(),
  buildQaDockerHarnessImage: vi.fn(),
  runQaDockerUp: vi.fn(),
-  defaultQaRuntimeModelForMode: vi.fn<(mode: string, options?: { alternate?: boolean }) => string>(
-    (mode, options) =>
-      mode === "live-frontier"
-        ? "openai/gpt-5.4"
-        : options?.alternate
-          ? "mock-openai/gpt-5.4-alt"
-          : "mock-openai/gpt-5.4",
-  ),
+  defaultQaRuntimeModelForMode:
+    vi.fn<(mode: string, options?: { alternate?: boolean }) => string>(),
 }));

 vi.mock("./manual-lane.runtime.js", () => ({
@@ -83,6 +77,8 @@ import {
  runQaSuiteCommand,
 } from "./cli.runtime.js";
 import { runQaTelegramCommand } from "./live-transports/telegram/cli.runtime.js";
+import { defaultQaModelForMode as defaultQaProviderModelForMode } from "./model-selection.js";
+import type { QaProviderModeInput } from "./run-config.js";

 describe("qa cli runtime", () => {
  let stdoutWrite: ReturnType<typeof vi.spyOn>;
@@ -100,11 +96,7 @@ describe("qa cli runtime", () => {
    runQaDockerUp.mockReset();
    defaultQaRuntimeModelForMode.mockImplementation(
      (mode: string, options?: { alternate?: boolean }) =>
-        mode === "live-frontier"
-          ? "openai/gpt-5.4"
-          : options?.alternate
-            ? "mock-openai/gpt-5.4-alt"
-            : "mock-openai/gpt-5.4",
+        defaultQaProviderModelForMode(mode as QaProviderModeInput, options),
    );
    runQaSuiteFromRuntime.mockResolvedValue({
      watchUrl: "http://127.0.0.1:43124",
@@ -234,22 +226,6 @@ describe("qa cli runtime", () => {
    );
  });

-  it("normalizes legacy live-openai suite runs onto the frontier provider mode", async () => {
-    await runQaSuiteCommand({
-      repoRoot: "/tmp/openclaw-repo",
-      providerMode: "live-openai",
-      scenarioIds: ["approval-turn-tool-followthrough"],
-    });
-
-    expect(runQaSuiteFromRuntime).toHaveBeenCalledWith(
-      expect.objectContaining({
-        repoRoot: path.resolve("/tmp/openclaw-repo"),
-        transportId: "qa-channel",
-        providerMode: "live-frontier",
-      }),
-    );
-  });
-
  it("passes host suite concurrency through", async () => {
    await runQaSuiteCommand({
      repoRoot: "/tmp/openclaw-repo",
@@ -602,6 +578,25 @@ describe("qa cli runtime", () => {
    });
  });

+  it("defaults manual aimock runs onto the aimock model lane", async () => {
+    await runQaManualLaneCommand({
+      repoRoot: "/tmp/openclaw-repo",
+      providerMode: "aimock",
+      message: "read qa kickoff and reply short",
+    });
+
+    expect(runQaManualLane).toHaveBeenCalledWith({
+      repoRoot: path.resolve("/tmp/openclaw-repo"),
+      transportId: "qa-channel",
+      providerMode: "aimock",
+      primaryModel: "aimock/gpt-5.4",
+      alternateModel: "aimock/gpt-5.4-alt",
+      fastMode: undefined,
+      message: "read qa kickoff and reply short",
+      timeoutMs: undefined,
+    });
+  });
+
  it("defaults manual frontier runs onto the frontier model lane", async () => {
    await runQaManualLaneCommand({
      repoRoot: "/tmp/openclaw-repo",
@@ -640,31 +635,11 @@ describe("qa cli runtime", () => {
    });
  });

-  it("normalizes legacy live-openai manual runs onto the frontier provider mode", async () => {
-    await runQaManualLaneCommand({
-      repoRoot: "/tmp/openclaw-repo",
-      providerMode: "live-openai",
-      message: "read qa kickoff and reply short",
-    });
-
-    expect(runQaManualLane).toHaveBeenCalledWith(
-      expect.objectContaining({
-        repoRoot: path.resolve("/tmp/openclaw-repo"),
-        transportId: "qa-channel",
-        providerMode: "live-frontier",
-        primaryModel: "openai/gpt-5.4",
-        alternateModel: "openai/gpt-5.4",
-      }),
-    );
-  });
-
  it("defaults manual frontier runs onto Codex OAuth when the runtime resolver prefers it", async () => {
    defaultQaRuntimeModelForMode.mockImplementation((mode, options) =>
      mode === "live-frontier"
        ? "openai-codex/gpt-5.4"
-        : options?.alternate
-          ? "mock-openai/gpt-5.4-alt"
-          : "mock-openai/gpt-5.4",
+        : defaultQaProviderModelForMode(mode as QaProviderModeInput, options),
    );

    await runQaManualLaneCommand({
--- a/extensions/qa-lab/src/cli.runtime.ts
+++ b/extensions/qa-lab/src/cli.runtime.ts
@@ -14,8 +14,13 @@ import { runQaDockerUp } from "./docker-up.runtime.js";
 import type { QaCliBackendAuthMode } from "./gateway-child.js";
 import { startQaLabServer } from "./lab-server.js";
 import { runQaManualLane } from "./manual-lane.runtime.js";
-import { startQaMockOpenAiServer } from "./mock-openai-server.js";
 import { runQaMultipass } from "./multipass.runtime.js";
+import { DEFAULT_QA_LIVE_PROVIDER_MODE, getQaProvider } from "./providers/index.js";
+import {
+  QA_FRONTIER_PARITY_BASELINE_LABEL,
+  QA_FRONTIER_PARITY_CANDIDATE_LABEL,
+} from "./providers/live-frontier/parity.js";
+import { startQaProviderServer } from "./providers/server-runtime.js";
 import {
  addQaCredentialSet,
  listQaCredentialSets,
@@ -419,8 +424,8 @@ export async function runQaParityReportCommand(opts: {
  ) as QaParitySuiteSummary;

  const comparison = buildQaAgenticParityComparison({
-    candidateLabel: opts.candidateLabel?.trim() || "openai/gpt-5.4",
-    baselineLabel: opts.baselineLabel?.trim() || "anthropic/claude-opus-4-6",
+    candidateLabel: opts.candidateLabel?.trim() || QA_FRONTIER_PARITY_CANDIDATE_LABEL,
+    baselineLabel: opts.baselineLabel?.trim() || QA_FRONTIER_PARITY_BASELINE_LABEL,
    candidateSummary,
    baselineSummary,
  });
@@ -488,7 +493,9 @@ export async function runQaManualLaneCommand(opts: {
  const repoRoot = path.resolve(opts.repoRoot ?? process.cwd());
  const transportId = normalizeQaTransportId(opts.transportId);
  const providerMode: QaProviderMode =
-    opts.providerMode === undefined ? "live-frontier" : normalizeQaProviderMode(opts.providerMode);
+    opts.providerMode === undefined
+      ? DEFAULT_QA_LIVE_PROVIDER_MODE
+      : normalizeQaProviderMode(opts.providerMode);
  const models = resolveQaManualLaneModels({
    providerMode,
    primaryModel: opts.primaryModel,
@@ -748,12 +755,23 @@ export async function runQaDockerUpCommand(opts: {
  process.stdout.write(`Stop: ${result.stopCommand}\n`);
 }

-export async function runQaMockOpenAiCommand(opts: { host?: string; port?: number }) {
-  const server = await startQaMockOpenAiServer({
+export async function runQaProviderServerCommand(
+  providerMode: QaProviderMode,
+  opts: { host?: string; port?: number },
+) {
+  const provider = getQaProvider(providerMode);
+  const standaloneCommand = provider.standaloneCommand;
+  if (!standaloneCommand) {
+    throw new Error(`QA provider "${providerMode}" does not expose a standalone server command.`);
+  }
+  const server = await startQaProviderServer(providerMode, {
    host: opts.host,
    port: Number.isFinite(opts.port) ? opts.port : undefined,
  });
-  await runInterruptibleServer("QA mock OpenAI", server);
+  if (!server) {
+    throw new Error(`QA provider "${providerMode}" does not expose a standalone server command.`);
+  }
+  await runInterruptibleServer(standaloneCommand.serverLabel, server);
 }

 export const __testing = {
--- a/extensions/qa-lab/src/cli.test.ts
+++ b/extensions/qa-lab/src/cli.test.ts
@@ -44,11 +44,13 @@ const {
  runQaCredentialsAddCommand,
  runQaCredentialsListCommand,
  runQaCredentialsRemoveCommand,
+  runQaProviderServerCommand,
  runQaTelegramCommand,
 } = vi.hoisted(() => ({
  runQaCredentialsAddCommand: vi.fn(),
  runQaCredentialsListCommand: vi.fn(),
  runQaCredentialsRemoveCommand: vi.fn(),
+  runQaProviderServerCommand: vi.fn(),
  runQaTelegramCommand: vi.fn(),
 }));

@@ -70,6 +72,7 @@ vi.mock("./cli.runtime.js", () => ({
  runQaCredentialsAddCommand,
  runQaCredentialsListCommand,
  runQaCredentialsRemoveCommand,
+  runQaProviderServerCommand,
 }));

 import { registerQaLabCli } from "./cli.js";
@@ -82,6 +85,7 @@ describe("qa cli registration", () => {
    runQaCredentialsAddCommand.mockReset();
    runQaCredentialsListCommand.mockReset();
    runQaCredentialsRemoveCommand.mockReset();
+    runQaProviderServerCommand.mockReset();
    runQaTelegramCommand.mockReset();
    listQaRunnerCliContributions
      .mockReset()
@@ -116,6 +120,20 @@ describe("qa cli registration", () => {
    );
  });

+  it("registers standalone provider server commands from the provider registry", async () => {
+    const qa = program.commands.find((command) => command.name() === "qa");
+    expect(qa?.commands.map((command) => command.name())).toEqual(
+      expect.arrayContaining(["mock-openai", "aimock"]),
+    );
+
+    await program.parseAsync(["node", "openclaw", "qa", "aimock", "--port", "44080"]);
+
+    expect(runQaProviderServerCommand).toHaveBeenCalledWith("aimock", {
+      host: "127.0.0.1",
+      port: 44080,
+    });
+  });
+
  it("shows an enable hint when a discovered runner plugin is installed but blocked", async () => {
    listQaRunnerCliContributions.mockReset().mockReturnValue([createBlockedQaRunnerContribution()]);
    const blockedProgram = new Command();
--- a/extensions/qa-lab/src/cli.ts
+++ b/extensions/qa-lab/src/cli.ts
@@ -1,7 +1,16 @@
 import type { Command } from "commander";
 import { collectString } from "./cli-options.js";
 import { listLiveTransportQaCliRegistrations } from "./live-transports/cli.js";
-import type { QaProviderModeInput } from "./run-config.js";
+import {
+  DEFAULT_QA_LIVE_PROVIDER_MODE,
+  formatQaProviderModeHelp,
+  listQaStandaloneProviderCommands,
+} from "./providers/index.js";
+import {
+  QA_FRONTIER_PARITY_BASELINE_LABEL,
+  QA_FRONTIER_PARITY_CANDIDATE_LABEL,
+} from "./providers/live-frontier/parity.js";
+import type { QaProviderMode, QaProviderModeInput } from "./run-config.js";
 import { hasQaScenarioPack } from "./scenario-catalog.js";

 type QaLabCliRuntime = typeof import("./cli.runtime.js");
@@ -174,9 +183,12 @@ async function runQaDockerUp(opts: {
  await runtime.runQaDockerUpCommand(opts);
 }

-async function runQaMockOpenAi(opts: { host?: string; port?: number }) {
+async function runQaProviderServer(
+  providerMode: QaProviderMode,
+  opts: { host?: string; port?: number },
+) {
  const runtime = await loadQaLabCliRuntime();
-  await runtime.runQaMockOpenAiCommand(opts);
+  await runtime.runQaProviderServerCommand(providerMode, opts);
 }

 export function isQaLabCliAvailable(): boolean {
@@ -208,11 +220,7 @@ export function registerQaLabCli(program: Command) {
    .option("--output-dir <path>", "Suite artifact directory")
    .option("--runner <kind>", "Execution runner: host or multipass", "host")
    .option("--transport <id>", "QA transport id", "qa-channel")
-    .option(
-      "--provider-mode <mode>",
-      "Provider mode: mock-openai or live-frontier (legacy live-openai still works)",
-      "live-frontier",
-    )
+    .option("--provider-mode <mode>", formatQaProviderModeHelp(), DEFAULT_QA_LIVE_PROVIDER_MODE)
    .option("--model <ref>", "Primary provider/model ref")
    .option("--alt-model <ref>", "Alternate provider/model ref")
    .option(
@@ -274,8 +282,12 @@ export function registerQaLabCli(program: Command) {
    .requiredOption("--candidate-summary <path>", "Candidate qa-suite-summary.json path")
    .requiredOption("--baseline-summary <path>", "Baseline qa-suite-summary.json path")
    .option("--repo-root <path>", "Repository root to target when running from a neutral cwd")
-    .option("--candidate-label <label>", "Candidate display label", "openai/gpt-5.4")
-    .option("--baseline-label <label>", "Baseline display label", "anthropic/claude-opus-4-6")
+    .option(
+      "--candidate-label <label>",
+      "Candidate display label",
+      QA_FRONTIER_PARITY_CANDIDATE_LABEL,
+    )
+    .option("--baseline-label <label>", "Baseline display label", QA_FRONTIER_PARITY_BASELINE_LABEL)
    .option("--output-dir <path>", "Artifact directory for the parity report")
    .action(
      async (opts: {
@@ -355,11 +367,7 @@ export function registerQaLabCli(program: Command) {
    .requiredOption("--message <text>", "Prompt to send to the QA agent")
    .option("--repo-root <path>", "Repository root to target when running from a neutral cwd")
    .option("--transport <id>", "QA transport id", "qa-channel")
-    .option(
-      "--provider-mode <mode>",
-      "Provider mode: mock-openai or live-frontier (legacy live-openai still works)",
-      "live-frontier",
-    )
+    .option("--provider-mode <mode>", formatQaProviderModeHelp(), DEFAULT_QA_LIVE_PROVIDER_MODE)
    .option("--model <ref>", "Primary provider/model ref (defaults by provider mode)")
    .option("--alt-model <ref>", "Alternate provider/model ref")
    .option("--fast", "Enable provider fast mode where supported", false)
@@ -574,13 +582,15 @@ export function registerQaLabCli(program: Command) {
      },
    );

-  qa.command("mock-openai")
-    .description("Run the local mock OpenAI Responses API server for QA")
-    .option("--host <host>", "Bind host", "127.0.0.1")
-    .option("--port <port>", "Bind port", (value: string) => Number(value))
-    .action(async (opts: { host?: string; port?: number }) => {
-      await runQaMockOpenAi(opts);
-    });
+  for (const providerCommand of listQaStandaloneProviderCommands()) {
+    qa.command(providerCommand.name)
+      .description(providerCommand.description)
+      .option("--host <host>", "Bind host", "127.0.0.1")
+      .option("--port <port>", "Bind port", (value: string) => Number(value))
+      .action(async (opts: { host?: string; port?: number }) => {
+        await runQaProviderServer(providerCommand.providerMode, opts);
+      });
+  }

  for (const lane of listLiveTransportQaCliRegistrations()) {
    assertNoQaSubcommandCollision(qa, lane.commandName);
--- a/extensions/qa-lab/src/gateway-child.ts
+++ b/extensions/qa-lab/src/gateway-child.ts
@@ -8,11 +8,6 @@ import path from "node:path";
 import { setTimeout as sleep } from "node:timers/promises";
 import type { OpenClawConfig } from "openclaw/plugin-sdk/config-runtime";
 import { formatErrorMessage } from "openclaw/plugin-sdk/error-runtime";
-import {
-  applyAuthProfileConfig,
-  upsertAuthProfile,
-  validateAnthropicSetupToken,
-} from "openclaw/plugin-sdk/provider-auth";
 import type { ModelProviderConfig } from "openclaw/plugin-sdk/provider-model-shared";
 import { fetchWithSsrFGuard } from "openclaw/plugin-sdk/ssrf-runtime";
 import { resolvePreferredOpenClawTmpDir } from "openclaw/plugin-sdk/temp-path";
@@ -25,72 +20,27 @@ import {
 import { assertRepoBoundPath, ensureRepoBoundDirectory } from "./cli-paths.js";
 import { formatQaGatewayLogsForError, redactQaGatewayDebugText } from "./gateway-log-redaction.js";
 import { startQaGatewayRpcClient } from "./gateway-rpc-client.js";
-import { splitQaModelRef } from "./model-selection.js";
+import { splitQaModelRef, type QaProviderMode } from "./model-selection.js";
 import { resolveQaNodeExecPath } from "./node-exec.js";
+import {
+  normalizeQaProviderModeEnv,
+  QA_LIVE_PROVIDER_CONFIG_PATH_ENV,
+  resolveQaLiveCliAuthEnv,
+  resolveQaLiveProviderConfigPath,
+  type QaCliBackendAuthMode,
+} from "./providers/env.js";
+import { DEFAULT_QA_PROVIDER_MODE, getQaProvider } from "./providers/index.js";
+import {
+  QA_LIVE_ANTHROPIC_SETUP_TOKEN_ENV,
+  QA_LIVE_SETUP_TOKEN_VALUE_ENV,
+  stageQaLiveAnthropicSetupToken,
+} from "./providers/live-frontier/auth.js";
+import { stageQaMockAuthProfiles } from "./providers/shared/mock-auth.js";
 import { seedQaAgentWorkspace } from "./qa-agent-workspace.js";
 import { buildQaGatewayConfig, type QaThinkingLevel } from "./qa-gateway-config.js";
 import type { QaTransportAdapter } from "./qa-transport.js";

-const QA_LIVE_ENV_ALIASES = Object.freeze([
-  {
-    liveVar: "OPENCLAW_LIVE_OPENAI_KEY",
-    providerVar: "OPENAI_API_KEY",
-  },
-  {
-    liveVar: "OPENCLAW_LIVE_ANTHROPIC_KEY",
-    providerVar: "ANTHROPIC_API_KEY",
-  },
-  {
-    liveVar: "OPENCLAW_LIVE_GEMINI_KEY",
-    providerVar: "GEMINI_API_KEY",
-  },
-]);
-
-const QA_MOCK_BLOCKED_ENV_VARS = Object.freeze([
-  "ANTHROPIC_API_KEY",
-  "ANTHROPIC_OAUTH_TOKEN",
-  "AWS_ACCESS_KEY_ID",
-  "AWS_BEARER_TOKEN_BEDROCK",
-  "AWS_REGION",
-  "AWS_SECRET_ACCESS_KEY",
-  "AWS_SESSION_TOKEN",
-  "GEMINI_API_KEY",
-  "GEMINI_API_KEYS",
-  "GOOGLE_API_KEY",
-  "MISTRAL_API_KEY",
-  "OPENAI_API_KEY",
-  "OPENAI_API_KEYS",
-  "OPENAI_BASE_URL",
-  "CODEX_HOME",
-  "OPENCLAW_LIVE_ANTHROPIC_KEY",
-  "OPENCLAW_LIVE_ANTHROPIC_KEYS",
-  "OPENCLAW_LIVE_GEMINI_KEY",
-  "OPENCLAW_LIVE_OPENAI_KEY",
-  "VOYAGE_API_KEY",
-]);
-
-const QA_MOCK_BLOCKED_ENV_KEY_PATTERNS = Object.freeze([
-  /^DISCORD_/i,
-  /^TELEGRAM_/i,
-  /^SLACK_/i,
-  /^MATRIX_/i,
-  /^SIGNAL_/i,
-  /^WHATSAPP_/i,
-  /^IMESSAGE_/i,
-  /^ZALO/i,
-  /^TWILIO_/i,
-  /^PLIVO_/i,
-  /^NGROK_/i,
-]);
-
-const QA_LIVE_PROVIDER_CONFIG_PATH_ENV = "OPENCLAW_QA_LIVE_PROVIDER_CONFIG_PATH";
-const QA_LIVE_ANTHROPIC_SETUP_TOKEN_ENV = "OPENCLAW_QA_LIVE_ANTHROPIC_SETUP_TOKEN";
-const QA_LIVE_SETUP_TOKEN_VALUE_ENV = "OPENCLAW_LIVE_SETUP_TOKEN_VALUE";
-const QA_LIVE_ANTHROPIC_SETUP_TOKEN_PROFILE_ENV = "OPENCLAW_QA_LIVE_ANTHROPIC_SETUP_TOKEN_PROFILE";
-const QA_LIVE_ANTHROPIC_SETUP_TOKEN_PROFILE_ID = "anthropic:qa-setup-token";
-const QA_LIVE_CLI_BACKEND_PRESERVE_ENV = "OPENCLAW_LIVE_CLI_BACKEND_PRESERVE_ENV";
-const QA_LIVE_CLI_BACKEND_AUTH_MODE_ENV = "OPENCLAW_LIVE_CLI_BACKEND_AUTH_MODE";
-export type QaCliBackendAuthMode = "auto" | "api-key" | "subscription";
+export type { QaCliBackendAuthMode } from "./providers/env.js";
 const QA_GATEWAY_CHILD_STARTUP_MAX_ATTEMPTS = 5;
 async function getFreePort() {
  return await new Promise<number>((resolve, reject) => {
@@ -199,108 +149,8 @@ function appendQaGatewayTempRoot(details: string, tempRoot: string) {
    : `${details}\nQA gateway temp root preserved at ${tempRoot}`;
 }

-export function normalizeQaProviderModeEnv(
-  env: NodeJS.ProcessEnv,
-  providerMode?: "mock-openai" | "live-frontier",
-) {
-  if (providerMode === "mock-openai") {
-    for (const key of QA_MOCK_BLOCKED_ENV_VARS) {
-      delete env[key];
-    }
-    for (const key of Object.keys(env)) {
-      if (QA_MOCK_BLOCKED_ENV_KEY_PATTERNS.some((pattern) => pattern.test(key))) {
-        delete env[key];
-      }
-    }
-    return env;
-  }
-
-  if (providerMode === "live-frontier") {
-    for (const { liveVar, providerVar } of QA_LIVE_ENV_ALIASES) {
-      const liveValue = env[liveVar]?.trim();
-      if (!liveValue || env[providerVar]?.trim()) {
-        continue;
-      }
-      env[providerVar] = liveValue;
-    }
-  }
-
-  return env;
-}
-
-export function resolveQaGatewayChildProviderMode(
-  providerMode?: "mock-openai" | "live-frontier",
-): "mock-openai" | "live-frontier" {
-  return providerMode ?? "mock-openai";
-}
-
-function resolveQaLiveCliAuthEnv(
-  baseEnv: NodeJS.ProcessEnv,
-  opts?: {
-    forwardHostHomeForClaudeCli?: boolean;
-    claudeCliAuthMode?: QaCliBackendAuthMode;
-  },
-) {
-  const parsePreservedCliEnv = () => {
-    const raw = baseEnv[QA_LIVE_CLI_BACKEND_PRESERVE_ENV]?.trim();
-    if (raw?.startsWith("[")) {
-      try {
-        const parsed = JSON.parse(raw) as unknown;
-        return Array.isArray(parsed)
-          ? parsed.filter((entry): entry is string => typeof entry === "string")
-          : [];
-      } catch {
-        return [];
-      }
-    }
-    return (raw ?? "").split(/[,\s]+/).filter((entry) => entry.length > 0);
-  };
-  const renderPreservedCliEnv = (values: string[]) => JSON.stringify([...new Set(values)]);
-  const authMode = opts?.claudeCliAuthMode ?? "auto";
-  const hasAnthropicKey = Boolean(
-    baseEnv.ANTHROPIC_API_KEY?.trim() || baseEnv.OPENCLAW_LIVE_ANTHROPIC_KEY?.trim(),
-  );
-  if (opts?.forwardHostHomeForClaudeCli && authMode === "api-key" && !hasAnthropicKey) {
-    throw new Error(
-      "Claude CLI API-key QA mode requires ANTHROPIC_API_KEY or OPENCLAW_LIVE_ANTHROPIC_KEY",
-    );
-  }
-  const preserveEnvValues = (() => {
-    if (!opts?.forwardHostHomeForClaudeCli) {
-      return undefined;
-    }
-    const values = parsePreservedCliEnv().filter((entry) => entry !== "ANTHROPIC_API_KEY");
-    if (authMode === "api-key" || (authMode === "auto" && hasAnthropicKey)) {
-      values.push("ANTHROPIC_API_KEY");
-    }
-    return renderPreservedCliEnv(values);
-  })();
-  const claudeCliEnv = opts?.forwardHostHomeForClaudeCli
-    ? {
-        [QA_LIVE_CLI_BACKEND_AUTH_MODE_ENV]: authMode,
-        ...(preserveEnvValues ? { [QA_LIVE_CLI_BACKEND_PRESERVE_ENV]: preserveEnvValues } : {}),
-      }
-    : {};
-  const configuredCodexHome = baseEnv.CODEX_HOME?.trim();
-  if (configuredCodexHome) {
-    return {
-      CODEX_HOME: configuredCodexHome,
-      ...claudeCliEnv,
-      ...(opts?.forwardHostHomeForClaudeCli && baseEnv.HOME?.trim()
-        ? { HOME: baseEnv.HOME.trim() }
-        : {}),
-    };
-  }
-  const hostHome = baseEnv.HOME?.trim();
-  if (!hostHome) {
-    return {};
-  }
-  const codexHome = path.join(hostHome, ".codex");
-  return {
-    ...(existsSync(codexHome) ? { CODEX_HOME: codexHome } : {}),
-    ...claudeCliEnv,
-    ...(opts?.forwardHostHomeForClaudeCli ? { HOME: hostHome } : {}),
-  };
+export function resolveQaGatewayChildProviderMode(providerMode?: QaProviderMode): QaProviderMode {
+  return providerMode ?? DEFAULT_QA_PROVIDER_MODE;
 }

 export function buildQaRuntimeEnv(params: {
@@ -314,19 +164,20 @@ export function buildQaRuntimeEnv(params: {
  xdgCacheHome: string;
  bundledPluginsDir?: string;
  compatibilityHostVersion?: string;
-  providerMode?: "mock-openai" | "live-frontier";
+  providerMode?: QaProviderMode;
  baseEnv?: NodeJS.ProcessEnv;
  forwardHostHomeForClaudeCli?: boolean;
  claudeCliAuthMode?: QaCliBackendAuthMode;
 }) {
  const baseEnv = params.baseEnv ?? process.env;
+  const provider = params.providerMode ? getQaProvider(params.providerMode) : null;
  const forwardedHostHome = params.forwardHostHome
    ? baseEnv.HOME?.trim() || os.homedir()
    : undefined;
  const env: NodeJS.ProcessEnv = {
    ...baseEnv,
    HOME: forwardedHostHome ?? params.homeDir,
-    ...(params.providerMode === "live-frontier"
+    ...(provider?.appliesLiveEnvAliases
      ? resolveQaLiveCliAuthEnv(baseEnv, {
          forwardHostHomeForClaudeCli: params.forwardHostHomeForClaudeCli,
          claudeCliAuthMode: params.claudeCliAuthMode,
@@ -360,119 +211,6 @@ export function buildQaRuntimeEnv(params: {
  return normalizedEnv;
 }

-function resolveQaLiveAnthropicSetupToken(env: NodeJS.ProcessEnv = process.env) {
-  const token = (
-    env[QA_LIVE_ANTHROPIC_SETUP_TOKEN_ENV]?.trim() ||
-    env[QA_LIVE_SETUP_TOKEN_VALUE_ENV]?.trim() ||
-    ""
-  ).replaceAll(/\s+/g, "");
-  if (!token) {
-    return null;
-  }
-  const tokenError = validateAnthropicSetupToken(token);
-  if (tokenError) {
-    throw new Error(`Invalid QA Anthropic setup-token: ${tokenError}`);
-  }
-  const profileId =
-    env[QA_LIVE_ANTHROPIC_SETUP_TOKEN_PROFILE_ENV]?.trim() ||
-    QA_LIVE_ANTHROPIC_SETUP_TOKEN_PROFILE_ID;
-  return { token, profileId };
-}
-
-export async function stageQaLiveAnthropicSetupToken(params: {
-  cfg: OpenClawConfig;
-  stateDir: string;
-  env?: NodeJS.ProcessEnv;
-}): Promise<OpenClawConfig> {
-  const resolved = resolveQaLiveAnthropicSetupToken(params.env);
-  if (!resolved) {
-    return params.cfg;
-  }
-  const agentDir = path.join(params.stateDir, "agents", "main", "agent");
-  await fs.mkdir(agentDir, { recursive: true });
-  upsertAuthProfile({
-    profileId: resolved.profileId,
-    credential: {
-      type: "token",
-      provider: "anthropic",
-      token: resolved.token,
-    },
-    agentDir,
-  });
-  return applyAuthProfileConfig(params.cfg, {
-    profileId: resolved.profileId,
-    provider: "anthropic",
-    mode: "token",
-    displayName: "QA setup-token",
-  });
-}
-
-/** Providers the mock-openai harness stages placeholder credentials for. */
-export const QA_MOCK_AUTH_PROVIDERS = Object.freeze(["openai", "anthropic"] as const);
-
-/** Agent IDs the mock-openai harness stages credentials under. */
-export const QA_MOCK_AUTH_AGENT_IDS = Object.freeze(["main", "qa"] as const);
-
-export function buildQaMockProfileId(provider: string): string {
-  return `qa-mock-${provider}`;
-}
-
-/**
- * In mock-openai mode the qa suite runs against the embedded mock server
- * instead of a real provider API. The mock does not validate credentials, but
- * the agent auth layer still needs a matching `api_key` auth profile in
- * `auth-profiles.json` before it will route the request through
- * `providerBaseUrl`. Without this staging step, every scenario fails with
- * `FailoverError: No API key found for provider "openai"` before the mock
- * server ever sees a request.
- *
- * Stages a placeholder `api_key` profile per provider in each of the agent
- * dirs the qa suite uses (`main` for the runtime config, `qa` for scenario
- * runs) and returns a config with matching `auth.profiles` entries so the
- * runtime accepts the profile on the first lookup.
- *
- * The placeholder value `qa-mock-not-a-real-key` is intentionally not
- * shaped like a real API key (no `sk-` prefix that would trip secret
- * scanners). It only needs to be non-empty to pass the credential
- * serializer; anything beyond that is ignored by the mock.
- */
-export async function stageQaMockAuthProfiles(params: {
-  cfg: OpenClawConfig;
-  stateDir: string;
-  agentIds?: readonly string[];
-  providers?: readonly string[];
-}): Promise<OpenClawConfig> {
-  const agentIds = [...new Set(params.agentIds ?? QA_MOCK_AUTH_AGENT_IDS)];
-  const providers = [...new Set(params.providers ?? QA_MOCK_AUTH_PROVIDERS)];
-  let next = params.cfg;
-  for (const agentId of agentIds) {
-    const agentDir = path.join(params.stateDir, "agents", agentId, "agent");
-    await fs.mkdir(agentDir, { recursive: true });
-    for (const provider of providers) {
-      const profileId = buildQaMockProfileId(provider);
-      upsertAuthProfile({
-        profileId,
-        credential: {
-          type: "api_key",
-          provider,
-          key: "qa-mock-not-a-real-key",
-          displayName: `QA mock ${provider} credential`,
-        },
-        agentDir,
-      });
-    }
-  }
-  for (const provider of providers) {
-    next = applyAuthProfileConfig(next, {
-      profileId: buildQaMockProfileId(provider),
-      provider,
-      mode: "api_key",
-      displayName: `QA mock ${provider} credential`,
-    });
-  }
-  return next;
-}
-
 function isRetryableGatewayCallError(details: string): boolean {
  return (
    details.includes("handshake timeout") ||
@@ -519,7 +257,6 @@ export const __testing = {
  redactQaGatewayDebugText,
  readQaLiveProviderConfigOverrides,
  resolveQaGatewayChildProviderMode,
-  resolveQaLiveAnthropicSetupToken,
  stageQaLiveAnthropicSetupToken,
  stageQaMockAuthProfiles,
  resolveQaLiveCliAuthEnv,
@@ -578,24 +315,6 @@ async function stopQaGatewayChildProcessTree(
  await waitForQaGatewayChildExit(child, opts?.forceTimeoutMs ?? 2_000);
 }

-function resolveQaUserPath(value: string, env: NodeJS.ProcessEnv = process.env) {
-  if (value === "~") {
-    return env.HOME ?? os.homedir();
-  }
-  if (value.startsWith("~/")) {
-    return path.join(env.HOME ?? os.homedir(), value.slice(2));
-  }
-  return path.resolve(value);
-}
-
-function resolveQaLiveProviderConfigPath(env: NodeJS.ProcessEnv = process.env) {
-  const explicit =
-    env[QA_LIVE_PROVIDER_CONFIG_PATH_ENV]?.trim() || env.OPENCLAW_CONFIG_PATH?.trim();
-  return explicit
-    ? { path: resolveQaUserPath(explicit, env), explicit: true }
-    : { path: path.join(os.homedir(), ".openclaw", "openclaw.json"), explicit: false };
-}
-
 function isRecord(value: unknown): value is Record<string, unknown> {
  return Boolean(value) && typeof value === "object" && !Array.isArray(value);
 }
@@ -705,7 +424,7 @@ export async function startQaGatewayChild(params: {
  transport: Pick<QaTransportAdapter, "requiredPluginIds" | "createGatewayConfig">;
  transportBaseUrl: string;
  controlUiAllowedOrigins?: string[];
-  providerMode?: "mock-openai" | "live-frontier";
+  providerMode?: QaProviderMode;
  primaryModel?: string;
  alternateModel?: string;
  fastMode?: boolean;
@@ -741,14 +460,14 @@ export async function startQaGatewayChild(params: {
    fs.mkdir(xdgCacheHome, { recursive: true }),
  ]);
  const providerMode = resolveQaGatewayChildProviderMode(params.providerMode);
-  const liveProviderIds =
-    providerMode === "live-frontier"
-      ? [params.primaryModel, params.alternateModel]
-          .map((modelRef) =>
-            typeof modelRef === "string" ? splitQaModelRef(modelRef)?.provider : undefined,
-          )
-          .filter((providerId): providerId is string => Boolean(providerId))
-      : [];
+  const resolvedProvider = getQaProvider(providerMode);
+  const liveProviderIds = resolvedProvider.usesModelProviderPlugins
+    ? [params.primaryModel, params.alternateModel]
+        .map((modelRef) =>
+          typeof modelRef === "string" ? splitQaModelRef(modelRef)?.provider : undefined,
+        )
+        .filter((providerId): providerId is string => Boolean(providerId))
+    : [];
  const liveProviderConfigs = await readQaLiveProviderConfigOverrides({
    providerIds: liveProviderIds,
  });
@@ -794,10 +513,12 @@ export async function startQaGatewayChild(params: {
      cfg,
      stateDir,
    });
-    if (providerMode === "mock-openai") {
+    const mockAuthProviders = getQaProvider(providerMode).mockAuthProviders;
+    if (mockAuthProviders && mockAuthProviders.length > 0) {
      cfg = await stageQaMockAuthProfiles({
        cfg,
        stateDir,
+        providers: mockAuthProviders,
      });
    }
    return params.mutateConfig ? params.mutateConfig(cfg) : cfg;
--- a/extensions/qa-lab/src/gateway-log-redaction.ts
+++ b/extensions/qa-lab/src/gateway-log-redaction.ts
@@ -1,23 +1,8 @@
+import { QA_PROVIDER_SECRET_ENV_VARS } from "./providers/env.js";
+
 const QA_GATEWAY_DEBUG_SECRET_ENV_VARS = Object.freeze([
-  "ANTHROPIC_API_KEY",
-  "ANTHROPIC_OAUTH_TOKEN",
-  "AWS_ACCESS_KEY_ID",
-  "AWS_BEARER_TOKEN_BEDROCK",
-  "AWS_SECRET_ACCESS_KEY",
-  "AWS_SESSION_TOKEN",
-  "ANTHROPIC_API_KEYS",
-  "GEMINI_API_KEY",
-  "GEMINI_API_KEYS",
-  "GOOGLE_API_KEY",
-  "MISTRAL_API_KEY",
-  "OPENAI_API_KEY",
-  "OPENAI_API_KEYS",
+  ...QA_PROVIDER_SECRET_ENV_VARS,
  "OPENCLAW_GATEWAY_TOKEN",
-  "OPENCLAW_LIVE_ANTHROPIC_KEY",
-  "OPENCLAW_LIVE_ANTHROPIC_KEYS",
-  "OPENCLAW_LIVE_GEMINI_KEY",
-  "OPENCLAW_LIVE_OPENAI_KEY",
-  "VOYAGE_API_KEY",
 ]);

 export function redactQaGatewayDebugText(text: string) {
--- a/extensions/qa-lab/src/live-timeout.ts
+++ b/extensions/qa-lab/src/live-timeout.ts
@@ -1,41 +1,21 @@
+import type { QaProviderMode } from "./model-selection.js";
+import { getQaProvider } from "./providers/index.js";
+
 type QaLiveTimeoutProfile = {
-  providerMode: "mock-openai" | "live-frontier";
+  providerMode: QaProviderMode;
  primaryModel: string;
  alternateModel: string;
 };

-function isAnthropicModel(modelRef: string) {
-  return modelRef.startsWith("anthropic/");
-}
-
-function isOpenAiModel(modelRef: string) {
-  return modelRef.startsWith("openai/");
-}
-
-function isGptFiveModel(modelRef: string) {
-  return isOpenAiModel(modelRef) && modelRef.slice("openai/".length).startsWith("gpt-5");
-}
-
-function isClaudeOpusModel(modelRef: string) {
-  return isAnthropicModel(modelRef) && modelRef.includes("claude-opus");
-}
-
 export function resolveQaLiveTurnTimeoutMs(
  profile: QaLiveTimeoutProfile,
  fallbackMs: number,
  modelRef = profile.primaryModel,
 ) {
-  if (profile.providerMode === "mock-openai") {
-    return fallbackMs;
-  }
-  if (isClaudeOpusModel(modelRef)) {
-    return Math.max(fallbackMs, 240_000);
-  }
-  if (isAnthropicModel(modelRef)) {
-    return Math.max(fallbackMs, 180_000);
-  }
-  if (isGptFiveModel(modelRef)) {
-    return Math.max(fallbackMs, 360_000);
-  }
-  return Math.max(fallbackMs, 120_000);
+  return getQaProvider(profile.providerMode).resolveTurnTimeoutMs({
+    primaryModel: profile.primaryModel,
+    alternateModel: profile.alternateModel,
+    modelRef,
+    fallbackMs,
+  });
 }
--- a/extensions/qa-lab/src/live-transports/shared/live-gateway.runtime.test.ts
+++ b/extensions/qa-lab/src/live-transports/shared/live-gateway.runtime.test.ts
@@ -1,16 +1,16 @@
 import { afterEach, beforeEach, describe, expect, it, vi } from "vitest";

-const { startQaGatewayChild, startQaMockOpenAiServer } = vi.hoisted(() => ({
+const { startQaGatewayChild, startQaProviderServer } = vi.hoisted(() => ({
  startQaGatewayChild: vi.fn(),
-  startQaMockOpenAiServer: vi.fn(),
+  startQaProviderServer: vi.fn(),
 }));

 vi.mock("../../gateway-child.js", () => ({
  startQaGatewayChild,
 }));

-vi.mock("../../mock-openai-server.js", () => ({
-  startQaMockOpenAiServer,
+vi.mock("../../providers/server-runtime.js", () => ({
+  startQaProviderServer,
 }));

 import { startQaLiveLaneGateway } from "./live-gateway.runtime.js";
@@ -48,17 +48,21 @@ describe("startQaLiveLaneGateway", () => {
    gatewayCall.mockReset();
    mockStop.mockReset();
    startQaGatewayChild.mockReset();
-    startQaMockOpenAiServer.mockReset();
+    startQaProviderServer.mockReset();

    startQaGatewayChild.mockResolvedValue({
      call: gatewayCall,
      cfg: {},
      stop: gatewayStop,
    });
-    startQaMockOpenAiServer.mockResolvedValue({
-      baseUrl: "http://127.0.0.1:44080",
-      stop: mockStop,
-    });
+    startQaProviderServer.mockImplementation(async (providerMode: string) =>
+      providerMode === "mock-openai"
+        ? {
+            baseUrl: "http://127.0.0.1:44080",
+            stop: mockStop,
+          }
+        : null,
+    );
  });

  afterEach(() => {
@@ -76,10 +80,7 @@ describe("startQaLiveLaneGateway", () => {
      controlUiEnabled: false,
    });

-    expect(startQaMockOpenAiServer).toHaveBeenCalledWith({
-      host: "127.0.0.1",
-      port: 0,
-    });
+    expect(startQaProviderServer).toHaveBeenCalledWith("mock-openai");
    expect(startQaGatewayChild).toHaveBeenCalledWith(
      expect.objectContaining({
        transportBaseUrl: "http://127.0.0.1:43123",
@@ -104,7 +105,7 @@ describe("startQaLiveLaneGateway", () => {
      controlUiEnabled: false,
    });

-    expect(startQaMockOpenAiServer).not.toHaveBeenCalled();
+    expect(startQaProviderServer).toHaveBeenCalledWith("live-frontier");
    expect(startQaGatewayChild).toHaveBeenCalledWith(
      expect.objectContaining({
        transportBaseUrl: "http://127.0.0.1:43123",
--- a/extensions/qa-lab/src/live-transports/shared/live-gateway.runtime.ts
+++ b/extensions/qa-lab/src/live-transports/shared/live-gateway.runtime.ts
@@ -1,12 +1,13 @@
 import type { OpenClawConfig } from "openclaw/plugin-sdk/config-runtime";
 import { startQaGatewayChild, type QaCliBackendAuthMode } from "../../gateway-child.js";
-import { startQaMockOpenAiServer } from "../../mock-openai-server.js";
+import type { QaProviderMode } from "../../model-selection.js";
+import { startQaProviderServer } from "../../providers/server-runtime.js";
 import type { QaThinkingLevel } from "../../qa-gateway-config.js";
 import { appendLiveLaneIssue } from "./live-lane-helpers.js";

 async function stopQaLiveLaneResources(resources: {
  gateway: Awaited<ReturnType<typeof startQaGatewayChild>>;
-  mock: Awaited<ReturnType<typeof startQaMockOpenAiServer>> | null;
+  mock: { baseUrl: string; stop(): Promise<void> } | null;
 }) {
  const errors: string[] = [];
  try {
@@ -36,7 +37,7 @@ export async function startQaLiveLaneGateway(params: {
  };
  transportBaseUrl: string;
  controlUiAllowedOrigins?: string[];
-  providerMode: "mock-openai" | "live-frontier";
+  providerMode: QaProviderMode;
  primaryModel: string;
  alternateModel: string;
  fastMode?: boolean;
@@ -45,13 +46,7 @@ export async function startQaLiveLaneGateway(params: {
  controlUiEnabled?: boolean;
  mutateConfig?: (cfg: OpenClawConfig) => OpenClawConfig;
 }) {
-  const mock =
-    params.providerMode === "mock-openai"
-      ? await startQaMockOpenAiServer({
-          host: "127.0.0.1",
-          port: 0,
-        })
-      : null;
+  const mock = await startQaProviderServer(params.providerMode);
  try {
    const gateway = await startQaGatewayChild({
      repoRoot: params.repoRoot,
--- a/extensions/qa-lab/src/live-transports/shared/live-transport-cli.runtime.ts
+++ b/extensions/qa-lab/src/live-transports/shared/live-transport-cli.runtime.ts
@@ -1,5 +1,6 @@
 import path from "node:path";
 import { resolveRepoRelativeOutputDir } from "../../cli-paths.js";
+import { DEFAULT_QA_LIVE_PROVIDER_MODE } from "../../providers/index.js";
 import type { QaProviderMode } from "../../run-config.js";
 import { normalizeQaProviderMode } from "../../run-config.js";
 import type { LiveTransportQaCommandOptions } from "./live-transport-cli.js";
@@ -18,7 +19,7 @@ export function resolveLiveTransportQaRunOptions(
    ),
    providerMode:
      opts.providerMode === undefined
-        ? "live-frontier"
+        ? DEFAULT_QA_LIVE_PROVIDER_MODE
        : normalizeQaProviderMode(opts.providerMode),
    primaryModel: opts.primaryModel,
    alternateModel: opts.alternateModel,
--- a/extensions/qa-lab/src/live-transports/shared/live-transport-cli.ts
+++ b/extensions/qa-lab/src/live-transports/shared/live-transport-cli.ts
@@ -1,5 +1,6 @@
 import type { Command } from "commander";
 import { collectString } from "../../cli-options.js";
+import { DEFAULT_QA_LIVE_PROVIDER_MODE, formatQaProviderModeHelp } from "../../providers/index.js";
 import type { QaProviderModeInput } from "../../run-config.js";

 export type LiveTransportQaCommandOptions = {
@@ -78,11 +79,7 @@ export function registerLiveTransportQaCli(params: {
    .description(params.description)
    .option("--repo-root <path>", "Repository root to target when running from a neutral cwd")
    .option("--output-dir <path>", params.outputDirHelp)
-    .option(
-      "--provider-mode <mode>",
-      "Provider mode: mock-openai or live-frontier (legacy live-openai still works)",
-      "live-frontier",
-    )
+    .option("--provider-mode <mode>", formatQaProviderModeHelp(), DEFAULT_QA_LIVE_PROVIDER_MODE)
    .option("--model <ref>", "Primary provider/model ref")
    .option("--alt-model <ref>", "Alternate provider/model ref")
    .option("--scenario <id>", params.scenarioHelp, collectString, [])
--- a/extensions/qa-lab/src/live-transports/telegram/telegram-live.runtime.ts
+++ b/extensions/qa-lab/src/live-transports/telegram/telegram-live.runtime.ts
@@ -6,6 +6,7 @@ import { formatErrorMessage } from "openclaw/plugin-sdk/error-runtime";
 import { fetchWithSsrFGuard } from "openclaw/plugin-sdk/ssrf-runtime";
 import { z } from "zod";
 import { startQaGatewayChild } from "../../gateway-child.js";
+import { DEFAULT_QA_LIVE_PROVIDER_MODE } from "../../providers/index.js";
 import {
  defaultQaModelForMode,
  normalizeQaProviderMode,
@@ -858,7 +859,9 @@ export async function runTelegramQaLive(params: {
  };

  const runtimeEnv = credentialLease.payload;
-  const providerMode = normalizeQaProviderMode(params.providerMode ?? "live-frontier");
+  const providerMode = normalizeQaProviderMode(
+    params.providerMode ?? DEFAULT_QA_LIVE_PROVIDER_MODE,
+  );
  const primaryModel = params.primaryModel?.trim() || defaultQaModelForMode(providerMode);
  const alternateModel = params.alternateModel?.trim() || defaultQaModelForMode(providerMode, true);
  const sutAccountId = params.sutAccountId?.trim() || "sut";
--- a/extensions/qa-lab/src/manual-lane.runtime.test.ts
+++ b/extensions/qa-lab/src/manual-lane.runtime.test.ts
@@ -1,9 +1,9 @@
 import { afterEach, beforeEach, describe, expect, it, vi } from "vitest";

-const { startQaLabServer, startQaGatewayChild, startQaMockOpenAiServer } = vi.hoisted(() => ({
+const { startQaLabServer, startQaGatewayChild, startQaProviderServer } = vi.hoisted(() => ({
  startQaLabServer: vi.fn(),
  startQaGatewayChild: vi.fn(),
-  startQaMockOpenAiServer: vi.fn(),
+  startQaProviderServer: vi.fn(),
 }));

 vi.mock("./lab-server.js", () => ({
@@ -14,8 +14,8 @@ vi.mock("./gateway-child.js", () => ({
  startQaGatewayChild,
 }));

-vi.mock("./mock-openai-server.js", () => ({
-  startQaMockOpenAiServer,
+vi.mock("./providers/server-runtime.js", () => ({
+  startQaProviderServer,
 }));

 import { runQaManualLane } from "./manual-lane.runtime.js";
@@ -31,7 +31,7 @@ describe("runQaManualLane", () => {
    labStop.mockReset();
    startQaLabServer.mockReset();
    startQaGatewayChild.mockReset();
-    startQaMockOpenAiServer.mockReset();
+    startQaProviderServer.mockReset();

    startQaLabServer.mockResolvedValue({
      listenUrl: "http://127.0.0.1:43124",
@@ -64,10 +64,14 @@ describe("runQaManualLane", () => {
      stop: gatewayStop,
    });

-    startQaMockOpenAiServer.mockResolvedValue({
-      baseUrl: "http://127.0.0.1:44080",
-      stop: mockStop,
-    });
+    startQaProviderServer.mockImplementation(async (providerMode: string) =>
+      providerMode === "mock-openai"
+        ? {
+            baseUrl: "http://127.0.0.1:44080",
+            stop: mockStop,
+          }
+        : null,
+    );
  });

  afterEach(() => {
@@ -84,10 +88,7 @@ describe("runQaManualLane", () => {
      timeoutMs: 5_000,
    });

-    expect(startQaMockOpenAiServer).toHaveBeenCalledWith({
-      host: "127.0.0.1",
-      port: 0,
-    });
+    expect(startQaProviderServer).toHaveBeenCalledWith("mock-openai");
    expect(startQaGatewayChild).toHaveBeenCalledWith(
      expect.objectContaining({
        repoRoot: "/tmp/openclaw-repo",
@@ -115,7 +116,7 @@ describe("runQaManualLane", () => {
      timeoutMs: 5_000,
    });

-    expect(startQaMockOpenAiServer).not.toHaveBeenCalled();
+    expect(startQaProviderServer).toHaveBeenCalledWith("live-frontier");
    expect(startQaLabServer).toHaveBeenCalledWith({
      repoRoot: "/tmp/openclaw-repo",
      embeddedGateway: "disabled",
--- a/extensions/qa-lab/src/manual-lane.runtime.ts
+++ b/extensions/qa-lab/src/manual-lane.runtime.ts
@@ -4,14 +4,15 @@ import { formatErrorMessage } from "openclaw/plugin-sdk/error-runtime";
 import { startQaGatewayChild } from "./gateway-child.js";
 import { startQaLabServer } from "./lab-server.js";
 import { resolveQaLiveTurnTimeoutMs } from "./live-timeout.js";
-import { startQaMockOpenAiServer } from "./mock-openai-server.js";
+import type { QaProviderMode } from "./model-selection.js";
+import { startQaProviderServer } from "./providers/server-runtime.js";
 import type { QaThinkingLevel } from "./qa-gateway-config.js";
 import { createQaTransportAdapter, type QaTransportId } from "./qa-transport-registry.js";

 type QaManualLaneParams = {
  repoRoot: string;
  transportId?: QaTransportId;
-  providerMode: "mock-openai" | "live-frontier";
+  providerMode: QaProviderMode;
  primaryModel: string;
  alternateModel: string;
  fastMode?: boolean;
@@ -21,7 +22,7 @@ type QaManualLaneParams = {
 };

 function resolveManualLaneTimeoutMs(params: {
-  providerMode: "mock-openai" | "live-frontier";
+  providerMode: QaProviderMode;
  primaryModel: string;
  alternateModel: string;
  timeoutMs?: number;
@@ -54,13 +55,7 @@ export async function runQaManualLane(params: QaManualLaneParams) {
    id: params.transportId ?? "qa-channel",
    state: lab.state,
  });
-  const mock =
-    params.providerMode === "mock-openai"
-      ? await startQaMockOpenAiServer({
-          host: "127.0.0.1",
-          port: 0,
-        })
-      : null;
+  const mock = await startQaProviderServer(params.providerMode);
  const gateway = await startQaGatewayChild({
    repoRoot: params.repoRoot,
    providerBaseUrl: mock ? `${mock.baseUrl}/v1` : undefined,
--- a/extensions/qa-lab/src/model-catalog.runtime.ts
+++ b/extensions/qa-lab/src/model-catalog.runtime.ts
@@ -2,14 +2,18 @@ import { spawn } from "node:child_process";
 import fs from "node:fs/promises";
 import path from "node:path";
 import { resolvePreferredOpenClawTmpDir } from "openclaw/plugin-sdk/temp-path";
+import { resolveQaNodeExecPath } from "./node-exec.js";
+import {
+  isPreferredQaLiveFrontierCatalogModel,
+  QA_FRONTIER_CATALOG_ALTERNATE_MODEL,
+  QA_FRONTIER_CATALOG_PRIMARY_MODEL,
+  QA_FRONTIER_PROVIDER_IDS,
+} from "./providers/live-frontier/catalog.js";
 import {
  createQaChannelGatewayConfig,
  QA_CHANNEL_REQUIRED_PLUGIN_IDS,
 } from "./qa-channel-transport.js";
 import { buildQaGatewayConfig } from "./qa-gateway-config.js";
-import { resolveQaNodeExecPath } from "./node-exec.js";
-
-const QA_FRONTIER_PROVIDER_IDS = ["anthropic", "google", "openai"] as const;

 type ModelRow = {
  key: string;
@@ -48,7 +52,7 @@ export function selectQaRunnerModelOptions(rows: ModelRow[]): QaRunnerModelOptio
        name: row.name,
        provider: parsed?.provider ?? "unknown",
        input: row.input,
-        preferred: row.key === "openai/gpt-5.4",
+        preferred: isPreferredQaLiveFrontierCatalogModel(row.key),
      } satisfies QaRunnerModelOption;
    });

@@ -110,8 +114,8 @@ export async function loadQaRunnerModelOptions(params: { repoRoot: string; signa
      gatewayToken: "qa-model-catalog",
      workspaceDir,
      providerMode: "live-frontier",
-      primaryModel: "openai/gpt-5.4",
-      alternateModel: "anthropic/claude-sonnet-4-6",
+      primaryModel: QA_FRONTIER_CATALOG_PRIMARY_MODEL,
+      alternateModel: QA_FRONTIER_CATALOG_ALTERNATE_MODEL,
      enabledProviderIds: [...QA_FRONTIER_PROVIDER_IDS],
      imageGenerationModel: null,
      controlUiEnabled: false,
@@ -128,24 +132,20 @@ export async function loadQaRunnerModelOptions(params: { repoRoot: string; signa
    await new Promise<void>((resolve, reject) => {
      let aborted = params.signal?.aborted === true;
      let forceKillTimer: NodeJS.Timeout | undefined;
-      const child = spawn(
-        nodeExecPath,
-        ["dist/index.js", "models", "list", "--all", "--json"],
-        {
-          cwd: params.repoRoot,
-          env: {
-            ...process.env,
-            HOME: homeDir,
-            OPENCLAW_HOME: homeDir,
-            OPENCLAW_CONFIG_PATH: configPath,
-            OPENCLAW_STATE_DIR: stateDir,
-            OPENCLAW_OAUTH_DIR: path.join(stateDir, "credentials"),
-            OPENCLAW_CODEX_DISCOVERY_LIVE: "0",
-          },
-          detached: process.platform !== "win32",
-          stdio: ["ignore", "pipe", "pipe"],
+      const child = spawn(nodeExecPath, ["dist/index.js", "models", "list", "--all", "--json"], {
+        cwd: params.repoRoot,
+        env: {
+          ...process.env,
+          HOME: homeDir,
+          OPENCLAW_HOME: homeDir,
+          OPENCLAW_CONFIG_PATH: configPath,
+          OPENCLAW_STATE_DIR: stateDir,
+          OPENCLAW_OAUTH_DIR: path.join(stateDir, "credentials"),
+          OPENCLAW_CODEX_DISCOVERY_LIVE: "0",
        },
-      );
+        detached: process.platform !== "win32",
+        stdio: ["ignore", "pipe", "pipe"],
+      });
      const cleanup = () => {
        params.signal?.removeEventListener("abort", abortCatalogLoad);
        if (forceKillTimer) {
--- a/extensions/qa-lab/src/model-selection.runtime.test.ts
+++ b/extensions/qa-lab/src/model-selection.runtime.test.ts
@@ -56,7 +56,7 @@ describe("qa model selection runtime", () => {
    expect(defaultQaRuntimeModelForMode("live-frontier")).toBe("openai/gpt-5.4");
  });

-  it("leaves mock-openai defaults unchanged", () => {
+  it("leaves mock defaults unchanged", () => {
    listProfilesForProvider.mockImplementation((_store: unknown, provider: string) =>
      provider === "openai-codex" ? ["openai-codex:user@example.com"] : [],
    );
@@ -65,5 +65,7 @@ describe("qa model selection runtime", () => {
    expect(defaultQaRuntimeModelForMode("mock-openai", { alternate: true })).toBe(
      "mock-openai/gpt-5.4-alt",
    );
+    expect(defaultQaRuntimeModelForMode("aimock")).toBe("aimock/gpt-5.4");
+    expect(defaultQaRuntimeModelForMode("aimock", { alternate: true })).toBe("aimock/gpt-5.4-alt");
  });
 });
--- a/extensions/qa-lab/src/model-selection.runtime.ts
+++ b/extensions/qa-lab/src/model-selection.runtime.ts
@@ -1,30 +1,13 @@
 import {
-  listProfilesForProvider,
-  loadAuthProfileStoreForRuntime,
-} from "openclaw/plugin-sdk/agent-runtime";
-import { resolveEnvApiKey } from "openclaw/plugin-sdk/provider-auth";
-import { defaultQaModelForMode, type QaProviderModeInput } from "./model-selection.js";
-
-const QA_CODEX_OAUTH_LIVE_MODEL = "openai-codex/gpt-5.4";
+  defaultQaModelForMode,
+  normalizeQaProviderMode,
+  type QaProviderModeInput,
+} from "./model-selection.js";
+import { DEFAULT_QA_LIVE_PROVIDER_MODE } from "./providers/index.js";
+import { resolveQaLiveFrontierPreferredModel } from "./providers/live-frontier/model-selection.runtime.js";

 export function resolveQaPreferredLiveModel() {
-  if (resolveEnvApiKey("openai")?.apiKey) {
-    return undefined;
-  }
-  try {
-    const store = loadAuthProfileStoreForRuntime(undefined, {
-      readOnly: true,
-      allowKeychainPrompt: false,
-    });
-    if (listProfilesForProvider(store, "openai").length > 0) {
-      return undefined;
-    }
-    return listProfilesForProvider(store, "openai-codex").length > 0
-      ? QA_CODEX_OAUTH_LIVE_MODEL
-      : undefined;
-  } catch {
-    return undefined;
-  }
+  return resolveQaLiveFrontierPreferredModel();
 }

 export function defaultQaRuntimeModelForMode(
@@ -34,8 +17,13 @@ export function defaultQaRuntimeModelForMode(
    preferredLiveModel?: string;
  },
 ) {
+  const preferredLiveModel =
+    options?.preferredLiveModel ??
+    (normalizeQaProviderMode(mode) === DEFAULT_QA_LIVE_PROVIDER_MODE
+      ? resolveQaPreferredLiveModel()
+      : undefined);
  return defaultQaModelForMode(mode, {
    ...options,
-    preferredLiveModel: options?.preferredLiveModel ?? resolveQaPreferredLiveModel(),
+    preferredLiveModel,
  });
 }
--- a/extensions/qa-lab/src/model-selection.ts
+++ b/extensions/qa-lab/src/model-selection.ts
@@ -1,14 +1,18 @@
-export type QaProviderMode = "mock-openai" | "live-frontier";
-export type QaProviderModeInput = QaProviderMode | "live-openai";
+import {
+  DEFAULT_QA_LIVE_PROVIDER_MODE,
+  getQaProvider,
+  type QaProviderMode,
+  type QaProviderModeInput,
+} from "./providers/index.js";
+
+export type { QaProviderMode, QaProviderModeInput } from "./providers/index.js";

 export type QaModelSelection = {
  primaryModel: string;
  alternateModel: string;
 };

-export function normalizeQaProviderMode(mode: QaProviderModeInput): QaProviderMode {
-  return mode === "live-openai" ? "live-frontier" : mode;
-}
+export { normalizeQaProviderMode } from "./providers/index.js";

 export function defaultQaModelForMode(
  mode: QaProviderModeInput,
@@ -17,10 +21,7 @@ export function defaultQaModelForMode(
    preferredLiveModel?: string;
  },
 ) {
-  if (normalizeQaProviderMode(mode) === "live-frontier") {
-    return options?.preferredLiveModel ?? "openai/gpt-5.4";
-  }
-  return options?.alternate ? "mock-openai/gpt-5.4-alt" : "mock-openai/gpt-5.4";
+  return getQaProvider(mode).defaultModel(options);
 }

 export function splitQaModelRef(ref: string) {
@@ -35,7 +36,7 @@ export function splitQaModelRef(ref: string) {
 }

 export function isQaFastModeModelRef(ref: string) {
-  return splitQaModelRef(ref)?.provider === "openai";
+  return getQaProvider(DEFAULT_QA_LIVE_PROVIDER_MODE).usesFastModeByDefault(ref);
 }

 export function isQaFastModeEnabled(selection: QaModelSelection) {
--- a/extensions/qa-lab/src/multipass.runtime.ts
+++ b/extensions/qa-lab/src/multipass.runtime.ts
@@ -2,9 +2,11 @@ import { execFile } from "node:child_process";
 import { randomUUID } from "node:crypto";
 import fs from "node:fs";
 import { access, appendFile, mkdir, writeFile } from "node:fs/promises";
-import os from "node:os";
 import path from "node:path";
 import { resolvePreferredOpenClawTmpDir } from "openclaw/plugin-sdk/temp-path";
+import type { QaProviderMode } from "./model-selection.js";
+import { resolveQaForwardedLiveEnv, resolveQaLiveProviderConfigPath } from "./providers/env.js";
+import { DEFAULT_QA_LIVE_PROVIDER_MODE, getQaProvider } from "./providers/index.js";

 const MULTIPASS_MOUNTED_REPO_PATH = "/workspace/openclaw-host";
 const MULTIPASS_GUEST_REPO_PATH = "/workspace/openclaw";
@@ -30,50 +32,6 @@ const MULTIPASS_REPO_SYNC_EXCLUDES = [
 const MULTIPASS_EXEC_MAX_BUFFER = 64 * 1024 * 1024;
 const MULTIPASS_GUEST_RUN_TIMEOUT_MS = 60 * 60 * 1000;

-const QA_LIVE_ENV_ALIASES = Object.freeze([
-  {
-    liveVar: "OPENCLAW_LIVE_OPENAI_KEY",
-    providerVar: "OPENAI_API_KEY",
-  },
-  {
-    liveVar: "OPENCLAW_LIVE_ANTHROPIC_KEY",
-    providerVar: "ANTHROPIC_API_KEY",
-  },
-  {
-    liveVar: "OPENCLAW_LIVE_GEMINI_KEY",
-    providerVar: "GEMINI_API_KEY",
-  },
-]);
-
-const QA_LIVE_ALLOWED_ENV_VARS = Object.freeze([
-  "ANTHROPIC_API_KEY",
-  "ANTHROPIC_OAUTH_TOKEN",
-  "AWS_ACCESS_KEY_ID",
-  "AWS_BEARER_TOKEN_BEDROCK",
-  "AWS_REGION",
-  "AWS_SECRET_ACCESS_KEY",
-  "AWS_SESSION_TOKEN",
-  "GEMINI_API_KEY",
-  "GEMINI_API_KEYS",
-  "GOOGLE_API_KEY",
-  "MISTRAL_API_KEY",
-  "OPENAI_API_KEY",
-  "OPENAI_API_KEYS",
-  "OPENAI_BASE_URL",
-  "OPENCLAW_LIVE_ANTHROPIC_KEY",
-  "OPENCLAW_LIVE_ANTHROPIC_KEYS",
-  "OPENCLAW_LIVE_GEMINI_KEY",
-  "OPENCLAW_LIVE_OPENAI_KEY",
-  "OPENCLAW_QA_LIVE_PROVIDER_CONFIG_PATH",
-  "OPENCLAW_CONFIG_PATH",
-  "VOYAGE_API_KEY",
-]);
-const QA_LIVE_ALLOWED_ENV_PATTERNS = Object.freeze([
-  /^[A-Z0-9_]+_API_KEYS$/u,
-  /^[A-Z0-9_]+_API_KEY_[0-9]+$/u,
-  /^OPENCLAW_LIVE_[A-Z0-9_]+_KEYS$/u,
-]);
-
 export const qaMultipassDefaultResources = {
  image: "lts",
  cpus: 2,
@@ -109,7 +67,7 @@ export type QaMultipassPlan = {
  disk: string;
  pnpmVersion: string;
  transportId: string;
-  providerMode: "mock-openai" | "live-frontier";
+  providerMode: QaProviderMode;
  primaryModel?: string;
  alternateModel?: string;
  fastMode?: boolean;
@@ -256,63 +214,6 @@ function resolveMultipassInstallHint() {
  return "https://multipass.run/install";
 }

-function resolveUserPath(value: string, env: NodeJS.ProcessEnv = process.env) {
-  if (value === "~") {
-    return env.HOME ?? os.homedir();
-  }
-  if (value.startsWith("~/")) {
-    return path.join(env.HOME ?? os.homedir(), value.slice(2));
-  }
-  return path.resolve(value);
-}
-
-function resolveLiveProviderConfigPath(env: NodeJS.ProcessEnv = process.env) {
-  const explicit =
-    env.OPENCLAW_QA_LIVE_PROVIDER_CONFIG_PATH?.trim() || env.OPENCLAW_CONFIG_PATH?.trim();
-  return explicit
-    ? { path: resolveUserPath(explicit, env), explicit: true }
-    : { path: path.join(os.homedir(), ".openclaw", "openclaw.json"), explicit: false };
-}
-
-function resolveQaLiveCliAuthEnv(baseEnv: NodeJS.ProcessEnv) {
-  const configuredCodexHome = baseEnv.CODEX_HOME?.trim();
-  if (configuredCodexHome) {
-    const codexHome = resolveUserPath(configuredCodexHome, baseEnv);
-    return fs.existsSync(codexHome) ? { CODEX_HOME: codexHome } : {};
-  }
-  const hostHome = baseEnv.HOME?.trim();
-  const effectiveHome = hostHome || os.homedir();
-  const codexHome = path.join(effectiveHome, ".codex");
-  return fs.existsSync(codexHome) ? { CODEX_HOME: codexHome } : {};
-}
-
-function resolveForwardedLiveEnv(baseEnv: NodeJS.ProcessEnv = process.env) {
-  const forwarded: Record<string, string> = {};
-  for (const [key, rawValue] of Object.entries(baseEnv)) {
-    if (
-      !QA_LIVE_ALLOWED_ENV_VARS.includes(key) &&
-      !QA_LIVE_ALLOWED_ENV_PATTERNS.some((pattern) => pattern.test(key))
-    ) {
-      continue;
-    }
-    const value = rawValue?.trim();
-    if (value) {
-      forwarded[key] = value;
-    }
-  }
-  for (const { liveVar, providerVar } of QA_LIVE_ENV_ALIASES) {
-    const liveValue = forwarded[liveVar]?.trim();
-    if (liveValue && !forwarded[providerVar]?.trim()) {
-      forwarded[providerVar] = liveValue;
-    }
-  }
-  const liveCliAuth = resolveQaLiveCliAuthEnv(baseEnv);
-  if (liveCliAuth.CODEX_HOME) {
-    forwarded.CODEX_HOME = liveCliAuth.CODEX_HOME;
-  }
-  return forwarded;
-}
-
 function createQaMultipassOutputDir(repoRoot: string) {
  return path.join(repoRoot, ".artifacts", "qa-e2e", `multipass-${createOutputStamp()}`);
 }
@@ -332,7 +233,7 @@ export function createQaMultipassPlan(params: {
  repoRoot: string;
  outputDir?: string;
  transportId?: string;
-  providerMode?: "mock-openai" | "live-frontier";
+  providerMode?: QaProviderMode;
  primaryModel?: string;
  alternateModel?: string;
  fastMode?: boolean;
@@ -346,11 +247,13 @@ export function createQaMultipassPlan(params: {
  const outputDir = params.outputDir ?? createQaMultipassOutputDir(params.repoRoot);
  const scenarioIds = [...new Set(params.scenarioIds ?? [])];
  const transportId = params.transportId?.trim() || "qa-channel";
-  const providerMode = params.providerMode ?? "live-frontier";
-  const forwardedEnv = providerMode === "live-frontier" ? resolveForwardedLiveEnv() : {};
+  const providerMode = params.providerMode ?? DEFAULT_QA_LIVE_PROVIDER_MODE;
+  const provider = getQaProvider(providerMode);
+  const forwardedEnv = provider.appliesLiveEnvAliases ? resolveQaForwardedLiveEnv() : {};
  const hostCodexHomePath = forwardedEnv.CODEX_HOME;
-  const liveProviderConfig =
-    providerMode === "live-frontier" ? resolveLiveProviderConfigPath() : undefined;
+  const liveProviderConfig = provider.usesModelProviderPlugins
+    ? resolveQaLiveProviderConfigPath()
+    : undefined;
  const hostLiveProviderConfigPath =
    liveProviderConfig && fs.existsSync(liveProviderConfig.path)
      ? liveProviderConfig.path
@@ -637,7 +540,7 @@ export async function runQaMultipass(params: {
  repoRoot: string;
  outputDir?: string;
  transportId?: string;
-  providerMode?: "mock-openai" | "live-frontier";
+  providerMode?: QaProviderMode;
  primaryModel?: string;
  alternateModel?: string;
  fastMode?: boolean;
--- a/extensions/qa-lab/src/providers/README.md
+++ b/extensions/qa-lab/src/providers/README.md
@@ -0,0 +1,21 @@
+# QA Provider Lanes
+
+QA provider lanes are registered in `index.ts` and implemented in one folder per
+provider. Shared provider contracts and mock-provider helpers live in `shared/`.
+Mock lanes should use `shared/mock-provider-definition.ts` unless they need a
+custom shape.
+
+Each provider definition owns:
+
+- its accepted `providerMode`
+- default primary, alternate, and image-generation model refs
+- gateway `models.providers` config, when the lane needs config injection
+- model runtime params such as SSE transport, fast mode, or thinking defaults
+- optional local server startup for mock lanes
+- optional placeholder auth profile providers for mock lanes
+- whether the lane uses real provider plugins and live environment aliases
+
+Shared suite code should import only `providers/index.ts` and ask the selected
+provider for behavior. Do not add provider-name branches to suite, gateway,
+manual-lane, or live-transport runtime code unless the registry contract is
+missing a needed capability.
--- a/extensions/qa-lab/src/providers/aimock/index.ts
+++ b/extensions/qa-lab/src/providers/aimock/index.ts
@@ -0,0 +1,9 @@
+import { createMockQaProviderDefinition } from "../shared/mock-provider-definition.js";
+
+export const aimockProviderDefinition = createMockQaProviderDefinition({
+  mode: "aimock",
+  commandName: "aimock",
+  commandDescription: "Run the local AIMock provider server for QA",
+  serverLabel: "QA AIMock",
+  mockAuthProviders: ["aimock", "openai", "anthropic"],
+});
--- a/extensions/qa-lab/src/providers/aimock/server.test.ts
+++ b/extensions/qa-lab/src/providers/aimock/server.test.ts
@@ -0,0 +1,82 @@
+import { describe, expect, it } from "vitest";
+import { startQaAimockServer } from "./server.js";
+
+function makeResponsesInput(text: string) {
+  return {
+    role: "user",
+    content: [
+      {
+        type: "input_text",
+        text,
+      },
+    ],
+  };
+}
+
+describe("qa aimock server", () => {
+  it("serves OpenAI Responses text replies and debug request snapshots", async () => {
+    const server = await startQaAimockServer({
+      host: "127.0.0.1",
+      port: 0,
+    });
+    try {
+      const response = await fetch(`${server.baseUrl}/v1/responses`, {
+        method: "POST",
+        headers: { "content-type": "application/json" },
+        body: JSON.stringify({
+          model: "aimock/gpt-5.4",
+          stream: false,
+          input: [makeResponsesInput("hello aimock")],
+        }),
+      });
+      expect(response.status).toBe(200);
+      expect(await response.json()).toMatchObject({
+        status: "completed",
+        model: "aimock/gpt-5.4",
+      });
+
+      const debug = await fetch(`${server.baseUrl}/debug/last-request`);
+      expect(debug.status).toBe(200);
+      expect(await debug.json()).toMatchObject({
+        prompt: "hello aimock",
+        allInputText: "hello aimock",
+        model: "aimock/gpt-5.4",
+        providerVariant: "openai",
+      });
+    } finally {
+      await server.stop();
+    }
+  });
+
+  it("records the request list for scenario assertions", async () => {
+    const server = await startQaAimockServer({
+      host: "127.0.0.1",
+      port: 0,
+    });
+    try {
+      const response = await fetch(`${server.baseUrl}/v1/responses`, {
+        method: "POST",
+        headers: { "content-type": "application/json" },
+        body: JSON.stringify({
+          model: "aimock/gpt-5.4",
+          stream: false,
+          input: [makeResponsesInput("@openclaw explain the QA lab")],
+        }),
+      });
+      expect(response.status).toBe(200);
+      expect(await response.json()).toMatchObject({
+        status: "completed",
+      });
+
+      const debug = await fetch(`${server.baseUrl}/debug/requests`);
+      expect(debug.status).toBe(200);
+      expect(await debug.json()).toEqual([
+        expect.objectContaining({
+          prompt: "@openclaw explain the QA lab",
+        }),
+      ]);
+    } finally {
+      await server.stop();
+    }
+  });
+});
--- a/extensions/qa-lab/src/providers/aimock/server.ts
+++ b/extensions/qa-lab/src/providers/aimock/server.ts
@@ -0,0 +1,203 @@
+import type { IncomingMessage, ServerResponse } from "node:http";
+import {
+  LLMock,
+  type ChatCompletionRequest,
+  type JournalEntry,
+  type Mountable,
+} from "@copilotkit/aimock";
+
+type AimockRequestSnapshot = {
+  raw: string;
+  body: Record<string, unknown>;
+  prompt: string;
+  allInputText: string;
+  toolOutput: string;
+  model: string;
+  providerVariant: "openai" | "anthropic" | "unknown";
+  imageInputCount: number;
+  plannedToolName?: string;
+};
+
+function writeJson(res: ServerResponse, status: number, body: unknown) {
+  const text = JSON.stringify(body);
+  res.writeHead(status, {
+    "content-type": "application/json; charset=utf-8",
+    "content-length": Buffer.byteLength(text),
+    "cache-control": "no-store",
+  });
+  res.end(text);
+}
+
+function stringifyContent(content: unknown): string {
+  if (typeof content === "string") {
+    return content;
+  }
+  if (Array.isArray(content)) {
+    return content
+      .map((part) => stringifyContent(part))
+      .filter(Boolean)
+      .join("\n");
+  }
+  if (content && typeof content === "object") {
+    const record = content as Record<string, unknown>;
+    if (typeof record.text === "string") {
+      return record.text;
+    }
+    if (typeof record.content === "string") {
+      return record.content;
+    }
+    if (typeof record.output === "string") {
+      return record.output;
+    }
+  }
+  return "";
+}
+
+function requestMessages(body: ChatCompletionRequest | null | undefined) {
+  return Array.isArray(body?.messages) ? body.messages : [];
+}
+
+function extractLastUserText(body: ChatCompletionRequest | null | undefined) {
+  const messages = requestMessages(body);
+  for (let index = messages.length - 1; index >= 0; index -= 1) {
+    const message = messages[index];
+    if (message?.role === "user") {
+      return stringifyContent(message.content);
+    }
+  }
+  return "";
+}
+
+function extractAllInputText(body: ChatCompletionRequest | null | undefined) {
+  return requestMessages(body)
+    .map((message) => stringifyContent(message.content))
+    .filter(Boolean)
+    .join("\n");
+}
+
+function extractToolOutput(body: ChatCompletionRequest | null | undefined) {
+  const messages = requestMessages(body);
+  for (let index = messages.length - 1; index >= 0; index -= 1) {
+    const message = messages[index];
+    if (message?.role === "tool") {
+      return stringifyContent(message.content);
+    }
+  }
+  return "";
+}
+
+function countImageInputs(value: unknown): number {
+  if (Array.isArray(value)) {
+    return value.reduce((sum, entry) => sum + countImageInputs(entry), 0);
+  }
+  if (!value || typeof value !== "object") {
+    return 0;
+  }
+  const record = value as Record<string, unknown>;
+  const type = typeof record.type === "string" ? record.type : "";
+  const imageLikeType =
+    type === "input_image" || type === "image" || type === "image_url" || type === "media";
+  const nested =
+    countImageInputs(record.content) +
+    countImageInputs(record.image_url) +
+    countImageInputs(record.source);
+  return (imageLikeType ? 1 : 0) + nested;
+}
+
+function resolveProviderVariant(model: string): AimockRequestSnapshot["providerVariant"] {
+  const normalized = model.trim().toLowerCase();
+  const provider = /^([^/:]+)[/:]/.exec(normalized)?.[1] ?? normalized;
+  if (provider === "openai" || provider === "aimock" || provider === "openai-codex") {
+    return "openai";
+  }
+  if (provider === "anthropic" || provider === "claude-cli") {
+    return "anthropic";
+  }
+  if (/^(?:gpt-|o1-|openai-)/.test(normalized)) {
+    return "openai";
+  }
+  if (/^(?:claude-|anthropic-)/.test(normalized)) {
+    return "anthropic";
+  }
+  return "unknown";
+}
+
+function extractPlannedToolName(entry: JournalEntry) {
+  const response = entry.response.fixture?.response as
+    | { toolCalls?: Array<{ name?: unknown }> }
+    | undefined;
+  const name = response?.toolCalls?.[0]?.name;
+  return typeof name === "string" && name.length > 0 ? name : undefined;
+}
+
+function toRequestSnapshot(entry: JournalEntry): AimockRequestSnapshot {
+  const body = entry.body ?? null;
+  const model = typeof body?.model === "string" ? body.model : "";
+  return {
+    raw: JSON.stringify(body ?? {}),
+    body: (body ?? {}) as Record<string, unknown>,
+    prompt: extractLastUserText(body),
+    allInputText: extractAllInputText(body),
+    toolOutput: extractToolOutput(body),
+    model,
+    providerVariant: resolveProviderVariant(model),
+    imageInputCount: countImageInputs(requestMessages(body)),
+    plannedToolName: extractPlannedToolName(entry),
+  };
+}
+
+function createDebugMount(mock: LLMock): Mountable {
+  return {
+    async handleRequest(_req: IncomingMessage, res: ServerResponse, pathname: string) {
+      const entries = mock.getRequests();
+      if (pathname === "/last-request") {
+        const lastEntry = entries.at(-1);
+        writeJson(
+          res,
+          200,
+          lastEntry ? toRequestSnapshot(lastEntry) : { ok: false, error: "no request recorded" },
+        );
+        return true;
+      }
+      if (pathname === "/requests") {
+        writeJson(
+          res,
+          200,
+          entries.map((entry) => toRequestSnapshot(entry)),
+        );
+        return true;
+      }
+      if (pathname === "/image-generations") {
+        writeJson(
+          res,
+          200,
+          entries
+            .filter((entry) => entry.path === "/v1/images/generations")
+            .map((entry) => entry.body ?? {}),
+        );
+        return true;
+      }
+      return false;
+    },
+  };
+}
+
+export async function startQaAimockServer(params?: { host?: string; port?: number }) {
+  const mock = new LLMock({
+    host: params?.host ?? "127.0.0.1",
+    port: params?.port ?? 0,
+    strict: false,
+    logLevel: "silent",
+  });
+
+  mock.mount("/debug", createDebugMount(mock));
+  mock.onMessage(/.*/, { content: "AIMOCK_QA_OK" });
+
+  await mock.start();
+  return {
+    baseUrl: mock.baseUrl,
+    async stop() {
+      await mock.stop();
+    },
+  };
+}
--- a/extensions/qa-lab/src/providers/env.ts
+++ b/extensions/qa-lab/src/providers/env.ts
@@ -0,0 +1,229 @@
+import { existsSync } from "node:fs";
+import os from "node:os";
+import path from "node:path";
+import type { QaProviderMode } from "./index.js";
+import { getQaProvider } from "./index.js";
+
+const QA_LIVE_ENV_ALIASES = Object.freeze([
+  {
+    liveVar: "OPENCLAW_LIVE_OPENAI_KEY",
+    providerVar: "OPENAI_API_KEY",
+  },
+  {
+    liveVar: "OPENCLAW_LIVE_ANTHROPIC_KEY",
+    providerVar: "ANTHROPIC_API_KEY",
+  },
+  {
+    liveVar: "OPENCLAW_LIVE_GEMINI_KEY",
+    providerVar: "GEMINI_API_KEY",
+  },
+]);
+
+export const QA_LIVE_PROVIDER_CONFIG_PATH_ENV = "OPENCLAW_QA_LIVE_PROVIDER_CONFIG_PATH";
+export const QA_LIVE_CLI_BACKEND_PRESERVE_ENV = "OPENCLAW_LIVE_CLI_BACKEND_PRESERVE_ENV";
+export const QA_LIVE_CLI_BACKEND_AUTH_MODE_ENV = "OPENCLAW_LIVE_CLI_BACKEND_AUTH_MODE";
+export type QaCliBackendAuthMode = "auto" | "api-key" | "subscription";
+
+export const QA_PROVIDER_SECRET_ENV_VARS = Object.freeze([
+  "ANTHROPIC_API_KEY",
+  "ANTHROPIC_OAUTH_TOKEN",
+  "AWS_ACCESS_KEY_ID",
+  "AWS_BEARER_TOKEN_BEDROCK",
+  "AWS_SECRET_ACCESS_KEY",
+  "AWS_SESSION_TOKEN",
+  "ANTHROPIC_API_KEYS",
+  "GEMINI_API_KEY",
+  "GEMINI_API_KEYS",
+  "GOOGLE_API_KEY",
+  "MISTRAL_API_KEY",
+  "OPENAI_API_KEY",
+  "OPENAI_API_KEYS",
+  "OPENCLAW_LIVE_ANTHROPIC_KEY",
+  "OPENCLAW_LIVE_ANTHROPIC_KEYS",
+  "OPENCLAW_LIVE_GEMINI_KEY",
+  "OPENCLAW_LIVE_OPENAI_KEY",
+  "VOYAGE_API_KEY",
+]);
+
+const QA_MOCK_BLOCKED_ENV_VARS = Object.freeze([
+  ...QA_PROVIDER_SECRET_ENV_VARS,
+  "AWS_REGION",
+  "OPENAI_BASE_URL",
+  "CODEX_HOME",
+]);
+
+const QA_MOCK_BLOCKED_ENV_KEY_PATTERNS = Object.freeze([
+  /^DISCORD_/i,
+  /^TELEGRAM_/i,
+  /^SLACK_/i,
+  /^MATRIX_/i,
+  /^SIGNAL_/i,
+  /^WHATSAPP_/i,
+  /^IMESSAGE_/i,
+  /^ZALO/i,
+  /^TWILIO_/i,
+  /^PLIVO_/i,
+  /^NGROK_/i,
+]);
+
+const QA_LIVE_ALLOWED_ENV_VARS = Object.freeze([
+  ...QA_PROVIDER_SECRET_ENV_VARS,
+  "AWS_REGION",
+  "OPENAI_BASE_URL",
+  QA_LIVE_PROVIDER_CONFIG_PATH_ENV,
+  "OPENCLAW_CONFIG_PATH",
+]);
+
+const QA_LIVE_ALLOWED_ENV_PATTERNS = Object.freeze([
+  /^[A-Z0-9_]+_API_KEYS$/u,
+  /^[A-Z0-9_]+_API_KEY_[0-9]+$/u,
+  /^OPENCLAW_LIVE_[A-Z0-9_]+_KEYS$/u,
+]);
+
+function resolveUserPath(value: string, env: NodeJS.ProcessEnv = process.env) {
+  if (value === "~") {
+    return env.HOME ?? os.homedir();
+  }
+  if (value.startsWith("~/")) {
+    return path.join(env.HOME ?? os.homedir(), value.slice(2));
+  }
+  return path.resolve(value);
+}
+
+function applyLiveProviderEnvAliases(env: NodeJS.ProcessEnv | Record<string, string>) {
+  for (const { liveVar, providerVar } of QA_LIVE_ENV_ALIASES) {
+    const liveValue = env[liveVar]?.trim();
+    if (!liveValue || env[providerVar]?.trim()) {
+      continue;
+    }
+    env[providerVar] = liveValue;
+  }
+}
+
+function parsePreservedCliEnv(baseEnv: NodeJS.ProcessEnv) {
+  const raw = baseEnv[QA_LIVE_CLI_BACKEND_PRESERVE_ENV]?.trim();
+  if (raw?.startsWith("[")) {
+    try {
+      const parsed = JSON.parse(raw) as unknown;
+      return Array.isArray(parsed)
+        ? parsed.filter((entry): entry is string => typeof entry === "string")
+        : [];
+    } catch {
+      return [];
+    }
+  }
+  return (raw ?? "").split(/[,\s]+/).filter((entry) => entry.length > 0);
+}
+
+function renderPreservedCliEnv(values: string[]) {
+  return JSON.stringify([...new Set(values)]);
+}
+
+export function normalizeQaProviderModeEnv(env: NodeJS.ProcessEnv, providerMode?: QaProviderMode) {
+  const provider = providerMode ? getQaProvider(providerMode) : null;
+  if (provider?.scrubsLiveProviderEnv) {
+    for (const key of QA_MOCK_BLOCKED_ENV_VARS) {
+      delete env[key];
+    }
+    for (const key of Object.keys(env)) {
+      if (QA_MOCK_BLOCKED_ENV_KEY_PATTERNS.some((pattern) => pattern.test(key))) {
+        delete env[key];
+      }
+    }
+    return env;
+  }
+
+  if (provider?.appliesLiveEnvAliases) {
+    applyLiveProviderEnvAliases(env);
+  }
+
+  return env;
+}
+
+export function resolveQaLiveCliAuthEnv(
+  baseEnv: NodeJS.ProcessEnv,
+  opts?: {
+    forwardHostHomeForClaudeCli?: boolean;
+    claudeCliAuthMode?: QaCliBackendAuthMode;
+  },
+) {
+  const authMode = opts?.claudeCliAuthMode ?? "auto";
+  const hasAnthropicKey = Boolean(
+    baseEnv.ANTHROPIC_API_KEY?.trim() || baseEnv.OPENCLAW_LIVE_ANTHROPIC_KEY?.trim(),
+  );
+  if (opts?.forwardHostHomeForClaudeCli && authMode === "api-key" && !hasAnthropicKey) {
+    throw new Error(
+      "Claude CLI API-key QA mode requires ANTHROPIC_API_KEY or OPENCLAW_LIVE_ANTHROPIC_KEY",
+    );
+  }
+  const preserveEnvValues = (() => {
+    if (!opts?.forwardHostHomeForClaudeCli) {
+      return undefined;
+    }
+    const values = parsePreservedCliEnv(baseEnv).filter((entry) => entry !== "ANTHROPIC_API_KEY");
+    if (authMode === "api-key" || (authMode === "auto" && hasAnthropicKey)) {
+      values.push("ANTHROPIC_API_KEY");
+    }
+    return renderPreservedCliEnv(values);
+  })();
+  const claudeCliEnv = opts?.forwardHostHomeForClaudeCli
+    ? {
+        [QA_LIVE_CLI_BACKEND_AUTH_MODE_ENV]: authMode,
+        ...(preserveEnvValues ? { [QA_LIVE_CLI_BACKEND_PRESERVE_ENV]: preserveEnvValues } : {}),
+      }
+    : {};
+  const configuredCodexHome = baseEnv.CODEX_HOME?.trim();
+  if (configuredCodexHome) {
+    return {
+      CODEX_HOME: configuredCodexHome,
+      ...claudeCliEnv,
+      ...(opts?.forwardHostHomeForClaudeCli && baseEnv.HOME?.trim()
+        ? { HOME: baseEnv.HOME.trim() }
+        : {}),
+    };
+  }
+  const hostHome = baseEnv.HOME?.trim();
+  if (!hostHome) {
+    return {};
+  }
+  const codexHome = path.join(hostHome, ".codex");
+  return {
+    ...(existsSync(codexHome) ? { CODEX_HOME: codexHome } : {}),
+    ...claudeCliEnv,
+    ...(opts?.forwardHostHomeForClaudeCli ? { HOME: hostHome } : {}),
+  };
+}
+
+export function resolveQaLiveProviderConfigPath(env: NodeJS.ProcessEnv = process.env) {
+  const explicit =
+    env[QA_LIVE_PROVIDER_CONFIG_PATH_ENV]?.trim() || env.OPENCLAW_CONFIG_PATH?.trim();
+  return explicit
+    ? { path: resolveUserPath(explicit, env), explicit: true }
+    : { path: path.join(os.homedir(), ".openclaw", "openclaw.json"), explicit: false };
+}
+
+export function resolveQaForwardedLiveEnv(baseEnv: NodeJS.ProcessEnv = process.env) {
+  const forwarded: Record<string, string> = {};
+  for (const [key, rawValue] of Object.entries(baseEnv)) {
+    if (
+      !QA_LIVE_ALLOWED_ENV_VARS.includes(key) &&
+      !QA_LIVE_ALLOWED_ENV_PATTERNS.some((pattern) => pattern.test(key))
+    ) {
+      continue;
+    }
+    const value = rawValue?.trim();
+    if (value) {
+      forwarded[key] = value;
+    }
+  }
+  applyLiveProviderEnvAliases(forwarded);
+
+  const configuredCodexHome = baseEnv.CODEX_HOME?.trim();
+  const codexHome = configuredCodexHome
+    ? resolveUserPath(configuredCodexHome, baseEnv)
+    : path.join(baseEnv.HOME?.trim() || os.homedir(), ".codex");
+  if (existsSync(codexHome)) {
+    forwarded.CODEX_HOME = codexHome;
+  }
+  return forwarded;
+}
--- a/extensions/qa-lab/src/providers/image-generation.test.ts
+++ b/extensions/qa-lab/src/providers/image-generation.test.ts
@@ -0,0 +1,46 @@
+import { describe, expect, it } from "vitest";
+import { buildQaImageGenerationConfigPatch } from "./image-generation.js";
+
+describe("QA provider image generation config", () => {
+  it("uses the selected mock provider for mock-openai image generation", () => {
+    const patch = buildQaImageGenerationConfigPatch({
+      providerMode: "mock-openai",
+      providerBaseUrl: "http://127.0.0.1:44080/v1",
+      requiredPluginIds: ["qa-channel"],
+    });
+
+    expect(patch.plugins.allow).toEqual(["memory-core", "qa-channel"]);
+    expect(patch.agents.defaults.imageGenerationModel.primary).toBe("mock-openai/gpt-image-1");
+    expect(patch.models?.providers["mock-openai"]?.baseUrl).toBe("http://127.0.0.1:44080/v1");
+  });
+
+  it("uses the selected mock provider for AIMock image generation", () => {
+    const patch = buildQaImageGenerationConfigPatch({
+      providerMode: "aimock",
+      providerBaseUrl: "http://127.0.0.1:45080/v1",
+      requiredPluginIds: [],
+    });
+
+    expect(patch.agents.defaults.imageGenerationModel.primary).toBe("aimock/gpt-image-1");
+    expect(patch.models?.providers.aimock?.baseUrl).toBe("http://127.0.0.1:45080/v1");
+    expect(patch.models?.providers["mock-openai"]).toBeUndefined();
+  });
+
+  it("enables the live image provider plugin without replacing live model config", () => {
+    const patch = buildQaImageGenerationConfigPatch({
+      providerMode: "live-frontier",
+      requiredPluginIds: ["qa-channel"],
+    });
+
+    expect(patch.plugins).toEqual({
+      allow: ["memory-core", "openai", "qa-channel"],
+      entries: {
+        openai: {
+          enabled: true,
+        },
+      },
+    });
+    expect(patch.agents.defaults.imageGenerationModel.primary).toBe("openai/gpt-image-1");
+    expect(patch).not.toHaveProperty("models");
+  });
+});
--- a/extensions/qa-lab/src/providers/image-generation.ts
+++ b/extensions/qa-lab/src/providers/image-generation.ts
@@ -0,0 +1,73 @@
+import type { QaProviderMode } from "./index.js";
+import { getQaProvider } from "./index.js";
+
+type QaImageGenerationPatchInput = {
+  providerMode: QaProviderMode;
+  providerBaseUrl?: string;
+  requiredPluginIds: readonly string[];
+};
+
+function splitModelProviderId(modelRef: string) {
+  const slash = modelRef.indexOf("/");
+  return slash > 0 ? modelRef.slice(0, slash) : null;
+}
+
+function uniqueNonEmpty(values: readonly (string | null | undefined)[]) {
+  return [
+    ...new Set(values.map((value) => value?.trim()).filter((value): value is string => !!value)),
+  ];
+}
+
+export function buildQaImageGenerationConfigPatch(input: QaImageGenerationPatchInput) {
+  const provider = getQaProvider(input.providerMode);
+  const imageModelRef = provider.defaultImageGenerationModel({
+    modelProviderIds: provider.defaultImageGenerationProviderIds,
+  });
+  if (!imageModelRef) {
+    throw new Error(
+      `QA provider "${input.providerMode}" does not expose an image generation model`,
+    );
+  }
+  const imageProviderId = splitModelProviderId(imageModelRef);
+  const modelPatch = (() => {
+    if (provider.kind !== "mock") {
+      return null;
+    }
+    if (!input.providerBaseUrl) {
+      throw new Error(`QA provider "${input.providerMode}" requires a mock provider URL`);
+    }
+    return provider.buildGatewayModels({
+      providerBaseUrl: input.providerBaseUrl,
+    });
+  })();
+  const providerPluginIds = provider.usesModelProviderPlugins ? [imageProviderId] : [];
+  const enabledPluginIds = uniqueNonEmpty(providerPluginIds);
+
+  return {
+    plugins: {
+      allow: uniqueNonEmpty(["memory-core", ...enabledPluginIds, ...input.requiredPluginIds]),
+      ...(enabledPluginIds.length > 0
+        ? {
+            entries: Object.fromEntries(
+              enabledPluginIds.map((pluginId) => [pluginId, { enabled: true }]),
+            ),
+          }
+        : {}),
+    },
+    ...(modelPatch
+      ? {
+          models: {
+            mode: modelPatch.mode,
+            providers: modelPatch.providers,
+          },
+        }
+      : {}),
+    agents: {
+      defaults: {
+        imageGenerationModel: {
+          primary: imageModelRef,
+        },
+      },
+    },
+  };
+}
--- a/extensions/qa-lab/src/providers/index.ts
+++ b/extensions/qa-lab/src/providers/index.ts
@@ -0,0 +1,62 @@
+import { aimockProviderDefinition } from "./aimock/index.js";
+import { liveFrontierProviderDefinition } from "./live-frontier/index.js";
+import { mockOpenAiProviderDefinition } from "./mock-openai/index.js";
+import type { QaProviderDefinition, QaProviderMode, QaProviderModeInput } from "./shared/types.js";
+
+export type {
+  QaMockProviderServer,
+  QaProviderDefinition,
+  QaProviderMode,
+  QaProviderModeInput,
+} from "./shared/types.js";
+
+const PROVIDERS = [
+  mockOpenAiProviderDefinition,
+  aimockProviderDefinition,
+  liveFrontierProviderDefinition,
+] as const satisfies readonly QaProviderDefinition[];
+
+export const DEFAULT_QA_PROVIDER_MODE = "mock-openai" satisfies QaProviderMode;
+export const DEFAULT_QA_LIVE_PROVIDER_MODE = "live-frontier" satisfies QaProviderMode;
+
+const PROVIDERS_BY_INPUT = new Map<QaProviderModeInput, QaProviderDefinition>();
+for (const provider of PROVIDERS) {
+  PROVIDERS_BY_INPUT.set(provider.mode, provider);
+}
+
+export function isQaProviderModeInput(input: unknown): input is QaProviderModeInput {
+  return typeof input === "string" && PROVIDERS_BY_INPUT.has(input as QaProviderModeInput);
+}
+
+export function normalizeQaProviderMode(input: QaProviderModeInput): QaProviderMode {
+  return getQaProvider(input).mode;
+}
+
+export function getQaProvider(input: QaProviderModeInput): QaProviderDefinition {
+  const provider = PROVIDERS_BY_INPUT.get(input);
+  if (!provider) {
+    throw new Error(`unknown QA provider mode: ${input}`);
+  }
+  return provider;
+}
+
+export function listQaProviderModes() {
+  return PROVIDERS.map((provider) => provider.mode);
+}
+
+export function formatQaProviderModeHelp() {
+  return `Provider mode: ${listQaProviderModes().join(", ")}`;
+}
+
+export function listQaStandaloneProviderCommands() {
+  return PROVIDERS.flatMap((provider) =>
+    provider.standaloneCommand
+      ? [
+          {
+            providerMode: provider.mode,
+            ...provider.standaloneCommand,
+          },
+        ]
+      : [],
+  );
+}
--- a/extensions/qa-lab/src/providers/live-frontier/auth.ts
+++ b/extensions/qa-lab/src/providers/live-frontier/auth.ts
@@ -0,0 +1,60 @@
+import fs from "node:fs/promises";
+import path from "node:path";
+import type { OpenClawConfig } from "openclaw/plugin-sdk/config-runtime";
+import {
+  applyAuthProfileConfig,
+  upsertAuthProfile,
+  validateAnthropicSetupToken,
+} from "openclaw/plugin-sdk/provider-auth";
+
+export const QA_LIVE_ANTHROPIC_SETUP_TOKEN_ENV = "OPENCLAW_QA_LIVE_ANTHROPIC_SETUP_TOKEN";
+export const QA_LIVE_SETUP_TOKEN_VALUE_ENV = "OPENCLAW_LIVE_SETUP_TOKEN_VALUE";
+const QA_LIVE_ANTHROPIC_SETUP_TOKEN_PROFILE_ENV = "OPENCLAW_QA_LIVE_ANTHROPIC_SETUP_TOKEN_PROFILE";
+const QA_LIVE_ANTHROPIC_SETUP_TOKEN_PROFILE_ID = "anthropic:qa-setup-token";
+
+function resolveQaLiveAnthropicSetupToken(env: NodeJS.ProcessEnv = process.env) {
+  const token = (
+    env[QA_LIVE_ANTHROPIC_SETUP_TOKEN_ENV]?.trim() ||
+    env[QA_LIVE_SETUP_TOKEN_VALUE_ENV]?.trim() ||
+    ""
+  ).replaceAll(/\s+/g, "");
+  if (!token) {
+    return null;
+  }
+  const tokenError = validateAnthropicSetupToken(token);
+  if (tokenError) {
+    throw new Error(`Invalid QA Anthropic setup-token: ${tokenError}`);
+  }
+  const profileId =
+    env[QA_LIVE_ANTHROPIC_SETUP_TOKEN_PROFILE_ENV]?.trim() ||
+    QA_LIVE_ANTHROPIC_SETUP_TOKEN_PROFILE_ID;
+  return { token, profileId };
+}
+
+export async function stageQaLiveAnthropicSetupToken(params: {
+  cfg: OpenClawConfig;
+  stateDir: string;
+  env?: NodeJS.ProcessEnv;
+}): Promise<OpenClawConfig> {
+  const resolved = resolveQaLiveAnthropicSetupToken(params.env);
+  if (!resolved) {
+    return params.cfg;
+  }
+  const agentDir = path.join(params.stateDir, "agents", "main", "agent");
+  await fs.mkdir(agentDir, { recursive: true });
+  upsertAuthProfile({
+    profileId: resolved.profileId,
+    credential: {
+      type: "token",
+      provider: "anthropic",
+      token: resolved.token,
+    },
+    agentDir,
+  });
+  return applyAuthProfileConfig(params.cfg, {
+    profileId: resolved.profileId,
+    provider: "anthropic",
+    mode: "token",
+    displayName: "QA setup-token",
+  });
+}
--- a/extensions/qa-lab/src/providers/live-frontier/catalog.ts
+++ b/extensions/qa-lab/src/providers/live-frontier/catalog.ts
@@ -0,0 +1,7 @@
+export const QA_FRONTIER_PROVIDER_IDS = ["anthropic", "google", "openai"] as const;
+export const QA_FRONTIER_CATALOG_PRIMARY_MODEL = "openai/gpt-5.4";
+export const QA_FRONTIER_CATALOG_ALTERNATE_MODEL = "anthropic/claude-sonnet-4-6";
+
+export function isPreferredQaLiveFrontierCatalogModel(modelRef: string) {
+  return modelRef === QA_FRONTIER_CATALOG_PRIMARY_MODEL;
+}
--- a/extensions/qa-lab/src/providers/live-frontier/character-eval.ts
+++ b/extensions/qa-lab/src/providers/live-frontier/character-eval.ts
@@ -0,0 +1,36 @@
+import type { QaThinkingLevel } from "../../qa-thinking.js";
+
+type QaFrontierCharacterModelOptions = {
+  thinkingDefault?: QaThinkingLevel;
+  fastMode?: boolean;
+};
+
+export const QA_FRONTIER_CHARACTER_EVAL_MODELS = Object.freeze([
+  "openai/gpt-5.4",
+  "openai/gpt-5.2",
+  "openai/gpt-5",
+  "anthropic/claude-opus-4-6",
+  "anthropic/claude-sonnet-4-6",
+  "zai/glm-5.1",
+  "moonshot/kimi-k2.5",
+  "google/gemini-3.1-pro-preview",
+]);
+
+export const QA_FRONTIER_CHARACTER_THINKING_BY_MODEL: Readonly<Record<string, QaThinkingLevel>> =
+  Object.freeze({
+    "openai/gpt-5.4": "xhigh",
+    "openai/gpt-5.2": "xhigh",
+    "openai/gpt-5": "xhigh",
+  });
+
+export const QA_FRONTIER_CHARACTER_JUDGE_MODELS = Object.freeze([
+  "openai/gpt-5.4",
+  "anthropic/claude-opus-4-6",
+]);
+
+export const QA_FRONTIER_CHARACTER_JUDGE_MODEL_OPTIONS: Readonly<
+  Record<string, QaFrontierCharacterModelOptions>
+> = Object.freeze({
+  "openai/gpt-5.4": { thinkingDefault: "xhigh" },
+  "anthropic/claude-opus-4-6": { thinkingDefault: "high" },
+});
--- a/extensions/qa-lab/src/providers/live-frontier/index.ts
+++ b/extensions/qa-lab/src/providers/live-frontier/index.ts
@@ -0,0 +1,61 @@
+import type { QaProviderDefinition } from "../shared/types.js";
+
+function isOpenAiModel(modelRef: string) {
+  return modelRef.startsWith("openai/");
+}
+
+function isAnthropicModel(modelRef: string) {
+  return modelRef.startsWith("anthropic/");
+}
+
+function isQaFastModeModelRef(modelRef: string) {
+  return isOpenAiModel(modelRef);
+}
+
+function isGptFiveModel(modelRef: string) {
+  return isOpenAiModel(modelRef) && modelRef.slice("openai/".length).startsWith("gpt-5");
+}
+
+function isClaudeOpusModel(modelRef: string) {
+  return isAnthropicModel(modelRef) && modelRef.includes("claude-opus");
+}
+
+export const liveFrontierProviderDefinition: QaProviderDefinition = {
+  mode: "live-frontier",
+  kind: "live",
+  defaultModel: (options) => options?.preferredLiveModel ?? "openai/gpt-5.4",
+  defaultImageGenerationProviderIds: ["openai"],
+  defaultImageGenerationModel: ({ modelProviderIds }) =>
+    modelProviderIds.includes("openai") ? "openai/gpt-image-1" : null,
+  usesFastModeByDefault: isQaFastModeModelRef,
+  resolveModelParams: ({ modelRef, fastMode, thinkingDefault }) => ({
+    transport: "sse",
+    openaiWsWarmup: false,
+    ...(fastMode === true || isQaFastModeModelRef(modelRef) ? { fastMode: true } : {}),
+    ...(thinkingDefault ? { thinking: thinkingDefault } : {}),
+  }),
+  resolveTurnTimeoutMs: ({ fallbackMs, modelRef }) => {
+    if (isClaudeOpusModel(modelRef)) {
+      return Math.max(fallbackMs, 240_000);
+    }
+    if (isAnthropicModel(modelRef)) {
+      return Math.max(fallbackMs, 180_000);
+    }
+    if (isGptFiveModel(modelRef)) {
+      return Math.max(fallbackMs, 360_000);
+    }
+    return Math.max(fallbackMs, 120_000);
+  },
+  buildGatewayModels: ({ liveProviderConfigs }) => {
+    const providers = liveProviderConfigs ?? {};
+    return Object.keys(providers).length > 0
+      ? {
+          mode: "merge",
+          providers,
+        }
+      : null;
+  },
+  usesModelProviderPlugins: true,
+  scrubsLiveProviderEnv: false,
+  appliesLiveEnvAliases: true,
+};
--- a/extensions/qa-lab/src/providers/live-frontier/model-selection.runtime.ts
+++ b/extensions/qa-lab/src/providers/live-frontier/model-selection.runtime.ts
@@ -0,0 +1,27 @@
+import {
+  listProfilesForProvider,
+  loadAuthProfileStoreForRuntime,
+} from "openclaw/plugin-sdk/agent-runtime";
+import { resolveEnvApiKey } from "openclaw/plugin-sdk/provider-auth";
+
+const QA_CODEX_OAUTH_LIVE_MODEL = "openai-codex/gpt-5.4";
+
+export function resolveQaLiveFrontierPreferredModel() {
+  if (resolveEnvApiKey("openai")?.apiKey) {
+    return undefined;
+  }
+  try {
+    const store = loadAuthProfileStoreForRuntime(undefined, {
+      readOnly: true,
+      allowKeychainPrompt: false,
+    });
+    if (listProfilesForProvider(store, "openai").length > 0) {
+      return undefined;
+    }
+    return listProfilesForProvider(store, "openai-codex").length > 0
+      ? QA_CODEX_OAUTH_LIVE_MODEL
+      : undefined;
+  } catch {
+    return undefined;
+  }
+}
--- a/extensions/qa-lab/src/providers/live-frontier/parity.ts
+++ b/extensions/qa-lab/src/providers/live-frontier/parity.ts
@@ -0,0 +1,2 @@
+export const QA_FRONTIER_PARITY_CANDIDATE_LABEL = "openai/gpt-5.4";
+export const QA_FRONTIER_PARITY_BASELINE_LABEL = "anthropic/claude-opus-4-6";
--- a/extensions/qa-lab/src/providers/mock-openai/index.ts
+++ b/extensions/qa-lab/src/providers/mock-openai/index.ts
@@ -0,0 +1,9 @@
+import { createMockQaProviderDefinition } from "../shared/mock-provider-definition.js";
+
+export const mockOpenAiProviderDefinition = createMockQaProviderDefinition({
+  mode: "mock-openai",
+  commandName: "mock-openai",
+  commandDescription: "Run the local mock OpenAI Responses API server for QA",
+  serverLabel: "QA mock OpenAI",
+  mockAuthProviders: ["openai", "anthropic"],
+});
--- a/extensions/qa-lab/src/providers/mock-openai/server.test.ts
+++ b/extensions/qa-lab/src/providers/mock-openai/server.test.ts
@@ -1,5 +1,5 @@
 import { afterEach, describe, expect, it } from "vitest";
-import { resolveProviderVariant, startQaMockOpenAiServer } from "./mock-openai-server.js";
+import { resolveProviderVariant, startQaMockOpenAiServer } from "./server.js";

 const cleanups: Array<() => Promise<void>> = [];
 const QA_IMAGE_PNG_BASE64 =
@@ -63,6 +63,23 @@ function makeUserInput(text: string) {
  };
 }

+const SESSIONS_SPAWN_TOOL = { type: "function", name: "sessions_spawn" } as const;
+const THREAD_SUBAGENT_CHILD_ERROR_TOKEN = "QA_SUBAGENT_CHILD_ERROR";
+const THREAD_SUBAGENT_TOOL_ERROR =
+  "thread=true requested but thread delivery is unavailable in this test harness.";
+
+function threadSubagentTask(token: string) {
+  return `Reply exactly \`${token}\`. This is the marker.`;
+}
+
+function explicitSessionsSpawnPrompt(token: string) {
+  return [
+    "Use sessions_spawn for this QA check.",
+    `task="${threadSubagentTask(token)}"`,
+    "label=qa-thread-subagent thread=true mode=session runTimeoutSeconds=30",
+  ].join(" ");
+}
+
 describe("qa mock openai server", () => {
  it("serves health and streamed responses", async () => {
    const server = await startQaMockOpenAiServer({
@@ -109,14 +126,14 @@ describe("qa mock openai server", () => {
      },
      body: JSON.stringify({
        stream: true,
-        input: [makeUserInput("Quiet streaming QA check: reply exactly `MATRIX_QA_STREAMING_OK`.")],
+        input: [makeUserInput("Quiet streaming QA check: reply exactly `QA_STREAMING_OK`.")],
      }),
    });
    expect(quietResponse.status).toBe(200);
    const quietBody = await quietResponse.text();
    expect(quietBody).toContain('"type":"response.output_text.delta"');
    expect(quietBody).toContain('"phase":"final_answer"');
-    expect(quietBody).toContain("MATRIX_QA_STREAMING_OK");
+    expect(quietBody).toContain("QA_STREAMING_OK");

    const blockResponse = await fetch(`${server.baseUrl}/v1/responses`, {
      method: "POST",
@@ -548,6 +565,7 @@ describe("qa mock openai server", () => {
      },
      body: JSON.stringify({
        stream: true,
+        tools: [SESSIONS_SPAWN_TOOL],
        input: [
          {
            role: "user",
@@ -568,6 +586,67 @@ describe("qa mock openai server", () => {
    expect(body).toContain('\\"thread\\":false');
  });

+  it("emits explicitly requested sessions_spawn tool calls", async () => {
+    const server = await startMockServer();
+
+    const body = await expectResponsesText(server, {
+      stream: true,
+      tools: [SESSIONS_SPAWN_TOOL],
+      input: [
+        {
+          role: "user",
+          content: [
+            {
+              type: "input_text",
+              text: explicitSessionsSpawnPrompt("QA_SUBAGENT_CHILD_FIXED"),
+            },
+          ],
+        },
+      ],
+    });
+    expect(body).toContain('"name":"sessions_spawn"');
+    expect(body).toContain('\\"label\\":\\"qa-thread-subagent\\"');
+    expect(body).toContain('\\"thread\\":true');
+    expect(body).toContain('\\"mode\\":\\"session\\"');
+    expect(body).toContain("QA_SUBAGENT_CHILD_FIXED");
+  });
+
+  it("surfaces sessions_spawn tool errors instead of echoing child-task markers", async () => {
+    const server = await startMockServer();
+
+    const body = await expectResponsesJson<{
+      output?: Array<{ content?: Array<{ text?: string }> }>;
+    }>(server, {
+      stream: false,
+      tools: [SESSIONS_SPAWN_TOOL],
+      input: [
+        makeUserInput(explicitSessionsSpawnPrompt(THREAD_SUBAGENT_CHILD_ERROR_TOKEN)),
+        {
+          type: "function_call",
+          name: "sessions_spawn",
+          arguments: JSON.stringify({
+            task: threadSubagentTask(THREAD_SUBAGENT_CHILD_ERROR_TOKEN),
+            label: "qa-thread-subagent",
+            thread: true,
+            mode: "session",
+            runTimeoutSeconds: 30,
+          }),
+        },
+        {
+          type: "function_call_output",
+          output: JSON.stringify({
+            status: "error",
+            error: THREAD_SUBAGENT_TOOL_ERROR,
+          }),
+        },
+      ],
+    });
+
+    const text = body.output?.[0]?.content?.[0]?.text ?? "";
+    expect(text).toContain(THREAD_SUBAGENT_TOOL_ERROR);
+    expect(text).not.toContain(THREAD_SUBAGENT_CHILD_ERROR_TOKEN);
+  });
+
  it("plans memory tools and serves mock image generations", async () => {
    const server = await startQaMockOpenAiServer({
      host: "127.0.0.1",
@@ -850,6 +929,7 @@ describe("qa mock openai server", () => {
      },
      body: JSON.stringify({
        stream: true,
+        tools: [SESSIONS_SPAWN_TOOL],
        input: [
          {
            role: "user",
@@ -875,6 +955,7 @@ describe("qa mock openai server", () => {
      },
      body: JSON.stringify({
        stream: true,
+        tools: [SESSIONS_SPAWN_TOOL],
        input: [
          {
            role: "user",
@@ -905,6 +986,7 @@ describe("qa mock openai server", () => {
      },
      body: JSON.stringify({
        stream: false,
+        tools: [SESSIONS_SPAWN_TOOL],
        input: [
          {
            role: "user",
@@ -961,6 +1043,7 @@ describe("qa mock openai server", () => {
      headers: { "content-type": "application/json" },
      body: JSON.stringify({
        stream: true,
+        tools: [SESSIONS_SPAWN_TOOL],
        input: [{ role: "user", content: [{ type: "input_text", text: prompt }] }],
      }),
    });
@@ -972,6 +1055,7 @@ describe("qa mock openai server", () => {
      headers: { "content-type": "application/json" },
      body: JSON.stringify({
        stream: true,
+        tools: [SESSIONS_SPAWN_TOOL],
        input: [{ role: "user", content: [{ type: "input_text", text: prompt }] }],
      }),
    });
@@ -1430,6 +1514,60 @@ describe("qa mock openai server", () => {
    });
  });

+  it("preserves Anthropic /v1/messages declared tools for explicit sessions_spawn prompts", async () => {
+    const server = await startMockServer();
+
+    const response = await fetch(`${server.baseUrl}/v1/messages`, {
+      method: "POST",
+      headers: { "content-type": "application/json" },
+      body: JSON.stringify({
+        model: "claude-opus-4-6",
+        max_tokens: 256,
+        tools: [
+          {
+            name: "sessions_spawn",
+            input_schema: { type: "object", properties: {} },
+          },
+        ],
+        messages: [
+          {
+            role: "user",
+            content: [
+              {
+                type: "text",
+                text: explicitSessionsSpawnPrompt("QA_SUBAGENT_CHILD_ANTHROPIC"),
+              },
+            ],
+          },
+        ],
+      }),
+    });
+    expect(response.status).toBe(200);
+    const body = (await response.json()) as {
+      stop_reason: string;
+      content: Array<Record<string, unknown>>;
+    };
+    expect(body.stop_reason).toBe("tool_use");
+    const toolUseBlock = body.content.find((block) => block.type === "tool_use") as
+      | { name: string; input: Record<string, unknown> }
+      | undefined;
+    expect(toolUseBlock?.name).toBe("sessions_spawn");
+    expect(toolUseBlock?.input).toMatchObject({
+      task: threadSubagentTask("QA_SUBAGENT_CHILD_ANTHROPIC"),
+      label: "qa-thread-subagent",
+      thread: true,
+      mode: "session",
+      runTimeoutSeconds: 30,
+    });
+
+    const debugResponse = await fetch(`${server.baseUrl}/debug/last-request`);
+    expect(debugResponse.status).toBe(200);
+    expect(await debugResponse.json()).toMatchObject({
+      model: "claude-opus-4-6",
+      plannedToolName: "sessions_spawn",
+    });
+  });
+
  it("dispatches Anthropic /v1/messages tool_result follow-ups through the shared scenario logic", async () => {
    // This verifies the Anthropic adapter correctly feeds tool_result
    // content blocks into the shared scenario dispatcher so downstream
--- a/extensions/qa-lab/src/providers/mock-openai/server.ts
+++ b/extensions/qa-lab/src/providers/mock-openai/server.ts
@@ -1,6 +1,6 @@
 import { createServer, type IncomingMessage, type ServerResponse } from "node:http";
 import { setTimeout as sleep } from "node:timers/promises";
-import { closeQaHttpServer } from "./bus-server.js";
+import { closeQaHttpServer } from "../../bus-server.js";

 type ResponsesInputItem = Record<string, unknown>;

@@ -142,8 +142,8 @@ const QA_REASONING_ONLY_RECOVERY_PROMPT_RE = /reasoning-only continuation qa che
 const QA_REASONING_ONLY_SIDE_EFFECT_PROMPT_RE = /reasoning-only after write safety check/i;
 const QA_EMPTY_RESPONSE_RECOVERY_PROMPT_RE = /empty response continuation qa check/i;
 const QA_EMPTY_RESPONSE_EXHAUSTION_PROMPT_RE = /empty response exhaustion qa check/i;
-const QA_QUIET_STREAMING_PROMPT_RE = /(?:matrix\s+)?quiet streaming qa check/i;
-const QA_BLOCK_STREAMING_PROMPT_RE = /(?:matrix\s+)?block streaming qa check/i;
+const QA_QUIET_STREAMING_PROMPT_RE = /quiet streaming qa check/i;
+const QA_BLOCK_STREAMING_PROMPT_RE = /block streaming qa check/i;
 const QA_REASONING_ONLY_RETRY_NEEDLE =
  "recorded reasoning but did not produce a user-visible answer";
 const QA_EMPTY_RESPONSE_RETRY_NEEDLE =
@@ -538,6 +538,77 @@ function extractLabeledMarkerDirective(text: string, label: string) {
  );
 }

+function extractQuotedToolArg(text: string, name: string) {
+  const escapedName = name.replaceAll(/[.*+?^${}()|[\]\\]/g, "\\$&");
+  return extractLastCapture(text, new RegExp(`\\b${escapedName}\\s*=\\s*"([^"]+)"`, "i"));
+}
+
+function extractBareToolArg(text: string, name: string) {
+  const escapedName = name.replaceAll(/[.*+?^${}()|[\]\\]/g, "\\$&");
+  return extractLastCapture(text, new RegExp(`\\b${escapedName}\\s*=\\s*([^\\s\\\`.,;:!?]+)`, "i"));
+}
+
+function hasDeclaredTool(body: Record<string, unknown>, name: string) {
+  const tools = Array.isArray(body.tools) ? body.tools : [];
+  return tools.some((tool) => {
+    if (!tool || typeof tool !== "object") {
+      return false;
+    }
+    const record = tool as Record<string, unknown>;
+    if (record.name === name) {
+      return true;
+    }
+    const nested = record.function;
+    return Boolean(
+      nested && typeof nested === "object" && (nested as { name?: unknown }).name === name,
+    );
+  });
+}
+
+function buildExplicitSessionsSpawnArgs(text: string): Record<string, unknown> | null {
+  if (!/\bsessions_spawn\b/i.test(text)) {
+    return null;
+  }
+  const task = extractQuotedToolArg(text, "task");
+  if (!task) {
+    return null;
+  }
+  const label = extractQuotedToolArg(text, "label") ?? extractBareToolArg(text, "label");
+  const mode = extractBareToolArg(text, "mode")?.toLowerCase();
+  const runTimeoutSecondsRaw = extractBareToolArg(text, "runTimeoutSeconds");
+  const runTimeoutSeconds =
+    runTimeoutSecondsRaw && /^\d+$/.test(runTimeoutSecondsRaw)
+      ? Number(runTimeoutSecondsRaw)
+      : undefined;
+  return {
+    task,
+    ...(label ? { label } : {}),
+    ...(extractBareToolArg(text, "thread")?.toLowerCase() === "true" ? { thread: true } : {}),
+    ...(mode === "session" || mode === "run" ? { mode } : {}),
+    ...(runTimeoutSeconds !== undefined ? { runTimeoutSeconds } : {}),
+  };
+}
+
+function extractToolErrorForNamedCall(params: {
+  allInputText: string;
+  input: ResponsesInputItem[];
+  name: string;
+  toolJson: Record<string, unknown> | null;
+}) {
+  const error = typeof params.toolJson?.error === "string" ? params.toolJson.error.trim() : "";
+  if (!error) {
+    return undefined;
+  }
+  const namedFunctionCall = params.input.some(
+    (item) => item.type === "function_call" && item.name === params.name,
+  );
+  const namedPromptReference = new RegExp(`\\b${params.name}\\b`, "i").test(params.allInputText);
+  if (namedFunctionCall || namedPromptReference) {
+    return error;
+  }
+  return undefined;
+}
+
 function isHeartbeatPrompt(text: string) {
  const trimmed = text.trim();
  if (!trimmed || /remember this fact/i.test(trimmed)) {
@@ -573,10 +644,19 @@ function buildAssistantText(
  const imageInputCount = countImageInputs(input);
  const activeMemorySummary = extractActiveMemorySummary(allInputText);
  const snackPreference = extractSnackPreference(activeMemorySummary ?? memorySnippet);
+  const sessionsSpawnError = extractToolErrorForNamedCall({
+    allInputText,
+    input,
+    name: "sessions_spawn",
+    toolJson,
+  });

  if (/what was the qa canary code/i.test(prompt) && rememberedFact) {
    return `Protocol note: the QA canary code was ${rememberedFact}.`;
  }
+  if (sessionsSpawnError) {
+    return `Protocol note: sessions_spawn failed: ${sessionsSpawnError}`;
+  }
  if (/remember this fact/i.test(prompt) && exactReplyDirective) {
    return exactReplyDirective;
  }
@@ -867,6 +947,7 @@ async function buildResponsesPayload(
  const isBaselineUnmentionedChannelChatter = /\bno bot ping here\b/i.test(prompt);
  const hasReasoningOnlyRetryInstruction = allInputText.includes(QA_REASONING_ONLY_RETRY_NEEDLE);
  const hasEmptyResponseRetryInstruction = allInputText.includes(QA_EMPTY_RESPONSE_RETRY_NEEDLE);
+  const canCallSessionsSpawn = hasDeclaredTool(body, "sessions_spawn");
  if (/remember this fact/i.test(prompt)) {
    return buildAssistantEvents(buildAssistantText(input, body, scenarioState));
  }
@@ -1110,7 +1191,7 @@ async function buildResponsesPayload(
      size: "1024x1024",
    });
  }
-  if (/subagent fanout synthesis check/i.test(prompt)) {
+  if (canCallSessionsSpawn && /subagent fanout synthesis check/i.test(prompt)) {
    if (!toolOutput && scenarioState.subagentFanoutPhase === 0) {
      scenarioState.subagentFanoutPhase = 1;
      return buildToolCallEventsWithArgs("sessions_spawn", {
@@ -1128,6 +1209,10 @@ async function buildResponsesPayload(
      });
    }
  }
+  const explicitSessionsSpawnArgs = buildExplicitSessionsSpawnArgs(allInputText);
+  if (canCallSessionsSpawn && explicitSessionsSpawnArgs && !toolOutput) {
+    return buildToolCallEventsWithArgs("sessions_spawn", explicitSessionsSpawnArgs);
+  }
  if (/tool continuity check/i.test(prompt) && !toolOutput) {
    return buildToolCallEventsWithArgs("read", { path: "QA_KICKOFF_TASK.md" });
  }
@@ -1155,7 +1240,11 @@ async function buildResponsesPayload(
      });
    }
  }
-  if ((/\bdelegate\b/i.test(prompt) || /subagent handoff/i.test(prompt)) && !toolOutput) {
+  if (
+    canCallSessionsSpawn &&
+    (/\bdelegate\b/i.test(prompt) || /subagent handoff/i.test(prompt)) &&
+    !toolOutput
+  ) {
    return buildToolCallEventsWithArgs("sessions_spawn", {
      task: "Inspect the QA workspace and return one concise protocol note.",
      label: "qa-sidecar",
@@ -1547,12 +1636,13 @@ async function buildMessagesPayload(
  const normalizedModel =
    typeof body.model === "string" && body.model.trim() !== "" ? body.model : "claude-opus-4-6";
  // Dispatch through the same scenario logic the /v1/responses route uses.
-  // The mock dispatcher only reads `body.input`, `body.model`, and
-  // `body.stream`, so a synthetic shim body is sufficient.
+  // Preserve declared tools so route-specific adapters mirror what the
+  // real provider request made available to the model.
  const dispatchBody: Record<string, unknown> = {
    input,
    model: normalizedModel,
    stream: false,
+    ...(Array.isArray(body.tools) ? { tools: body.tools } : {}),
  };
  const events = await buildResponsesPayload(dispatchBody, scenarioState);
  const extracted = extractFinalAssistantOutputFromEvents(events);
--- a/extensions/qa-lab/src/providers/server-runtime.ts
+++ b/extensions/qa-lab/src/providers/server-runtime.ts
@@ -0,0 +1,36 @@
+import { getQaProvider, type QaMockProviderServer, type QaProviderModeInput } from "./index.js";
+
+type QaProviderServerParams = {
+  host: string;
+  port: number;
+};
+
+async function startMockOpenAiProviderServer(params: QaProviderServerParams) {
+  const { startQaMockOpenAiServer } = await import("./mock-openai/server.js");
+  return await startQaMockOpenAiServer(params);
+}
+
+async function startAimockProviderServer(params: QaProviderServerParams) {
+  const { startQaAimockServer } = await import("./aimock/server.js");
+  return await startQaAimockServer(params);
+}
+
+export async function startQaProviderServer(
+  input: QaProviderModeInput,
+  params?: { host?: string; port?: number },
+): Promise<QaMockProviderServer | null> {
+  const provider = getQaProvider(input);
+  const serverParams = {
+    host: params?.host ?? "127.0.0.1",
+    port: params?.port ?? 0,
+  };
+  switch (provider.mode) {
+    case "mock-openai":
+      return await startMockOpenAiProviderServer(serverParams);
+    case "aimock":
+      return await startAimockProviderServer(serverParams);
+    case "live-frontier":
+    default:
+      return null;
+  }
+}
--- a/extensions/qa-lab/src/providers/shared/mock-auth.ts
+++ b/extensions/qa-lab/src/providers/shared/mock-auth.ts
@@ -0,0 +1,70 @@
+import fs from "node:fs/promises";
+import path from "node:path";
+import type { OpenClawConfig } from "openclaw/plugin-sdk/config-runtime";
+import { applyAuthProfileConfig, upsertAuthProfile } from "openclaw/plugin-sdk/provider-auth";
+
+/** Providers the mock harness stages placeholder credentials for by default. */
+export const QA_MOCK_AUTH_PROVIDERS = Object.freeze(["openai", "anthropic"] as const);
+
+/** Agent IDs the mock harness stages credentials under. */
+export const QA_MOCK_AUTH_AGENT_IDS = Object.freeze(["main", "qa"] as const);
+
+export function buildQaMockProfileId(provider: string): string {
+  return `qa-mock-${provider}`;
+}
+
+/**
+ * In mock provider modes the qa suite runs against an embedded mock server
+ * instead of a real provider API. The mock does not validate credentials, but
+ * the agent auth layer still needs a matching `api_key` auth profile in
+ * `auth-profiles.json` before it will route the request through
+ * `providerBaseUrl`. Without this staging step, every scenario fails with
+ * `FailoverError: No API key found for provider "openai"` before the mock
+ * server ever sees a request.
+ *
+ * Stages a placeholder `api_key` profile per provider in each of the agent
+ * dirs the qa suite uses (`main` for the runtime config, `qa` for scenario
+ * runs) and returns a config with matching `auth.profiles` entries so the
+ * runtime accepts the profile on the first lookup.
+ *
+ * The placeholder value `qa-mock-not-a-real-key` is intentionally not
+ * shaped like a real API key (no `sk-` prefix that would trip secret
+ * scanners). It only needs to be non-empty to pass the credential
+ * serializer; anything beyond that is ignored by the mock.
+ */
+export async function stageQaMockAuthProfiles(params: {
+  cfg: OpenClawConfig;
+  stateDir: string;
+  agentIds?: readonly string[];
+  providers?: readonly string[];
+}): Promise<OpenClawConfig> {
+  const agentIds = [...new Set(params.agentIds ?? QA_MOCK_AUTH_AGENT_IDS)];
+  const providers = [...new Set(params.providers ?? QA_MOCK_AUTH_PROVIDERS)];
+  let next = params.cfg;
+  for (const agentId of agentIds) {
+    const agentDir = path.join(params.stateDir, "agents", agentId, "agent");
+    await fs.mkdir(agentDir, { recursive: true });
+    for (const provider of providers) {
+      const profileId = buildQaMockProfileId(provider);
+      upsertAuthProfile({
+        profileId,
+        credential: {
+          type: "api_key",
+          provider,
+          key: "qa-mock-not-a-real-key",
+          displayName: `QA mock ${provider} credential`,
+        },
+        agentDir,
+      });
+    }
+  }
+  for (const provider of providers) {
+    next = applyAuthProfileConfig(next, {
+      profileId: buildQaMockProfileId(provider),
+      provider,
+      mode: "api_key",
+      displayName: `QA mock ${provider} credential`,
+    });
+  }
+  return next;
+}
--- a/extensions/qa-lab/src/providers/shared/mock-model-config.ts
+++ b/extensions/qa-lab/src/providers/shared/mock-model-config.ts
@@ -0,0 +1,104 @@
+import type { ModelProviderConfig } from "openclaw/plugin-sdk/provider-model-shared";
+
+const ZERO_COST = Object.freeze({
+  input: 0,
+  output: 0,
+  cacheRead: 0,
+  cacheWrite: 0,
+});
+
+function cloneProvider(provider: ModelProviderConfig): ModelProviderConfig {
+  return {
+    ...provider,
+    models: provider.models.map((model) => ({ ...model })),
+  };
+}
+
+export function trimTrailingApiV1(baseUrl: string) {
+  return baseUrl.replace(/\/v1\/?$/i, "");
+}
+
+export function createMockOpenAiResponsesProvider(baseUrl: string): ModelProviderConfig {
+  return {
+    baseUrl,
+    apiKey: "test",
+    api: "openai-responses",
+    request: {
+      allowPrivateNetwork: true,
+    },
+    models: [
+      {
+        id: "gpt-5.4",
+        name: "gpt-5.4",
+        api: "openai-responses",
+        reasoning: false,
+        input: ["text", "image"],
+        cost: ZERO_COST,
+        contextWindow: 128_000,
+        maxTokens: 4096,
+      },
+      {
+        id: "gpt-5.4-alt",
+        name: "gpt-5.4-alt",
+        api: "openai-responses",
+        reasoning: false,
+        input: ["text", "image"],
+        cost: ZERO_COST,
+        contextWindow: 128_000,
+        maxTokens: 4096,
+      },
+      {
+        id: "gpt-image-1",
+        name: "gpt-image-1",
+        api: "openai-responses",
+        reasoning: false,
+        input: ["text"],
+        cost: ZERO_COST,
+        contextWindow: 128_000,
+        maxTokens: 4096,
+      },
+    ],
+  };
+}
+
+export function createMockAnthropicMessagesProvider(baseUrl: string): ModelProviderConfig {
+  return {
+    baseUrl: trimTrailingApiV1(baseUrl),
+    apiKey: "test",
+    api: "anthropic-messages",
+    request: {
+      allowPrivateNetwork: true,
+    },
+    models: [
+      {
+        id: "claude-opus-4-6",
+        name: "claude-opus-4-6",
+        api: "anthropic-messages",
+        reasoning: false,
+        input: ["text", "image"],
+        cost: ZERO_COST,
+        contextWindow: 200_000,
+        maxTokens: 4096,
+      },
+      {
+        id: "claude-sonnet-4-6",
+        name: "claude-sonnet-4-6",
+        api: "anthropic-messages",
+        reasoning: false,
+        input: ["text", "image"],
+        cost: ZERO_COST,
+        contextWindow: 200_000,
+        maxTokens: 4096,
+      },
+    ],
+  };
+}
+
+export function createMockProviderMap(primaryProviderId: string, providerBaseUrl: string) {
+  const primaryProvider = createMockOpenAiResponsesProvider(providerBaseUrl);
+  return {
+    [primaryProviderId]: primaryProvider,
+    openai: cloneProvider(primaryProvider),
+    anthropic: createMockAnthropicMessagesProvider(providerBaseUrl),
+  };
+}
--- a/extensions/qa-lab/src/providers/shared/mock-provider-definition.ts
+++ b/extensions/qa-lab/src/providers/shared/mock-provider-definition.ts
@@ -0,0 +1,45 @@
+import { createMockProviderMap } from "./mock-model-config.js";
+import type { QaProviderDefinition, QaProviderMode } from "./types.js";
+
+export type MockQaProviderDefinitionParams = {
+  mode: Extract<QaProviderMode, "aimock" | "mock-openai">;
+  commandName: string;
+  commandDescription: string;
+  serverLabel: string;
+  mockAuthProviders: readonly string[];
+};
+
+function mockModelRef(providerId: string, alternate?: boolean) {
+  return `${providerId}/${alternate ? "gpt-5.4-alt" : "gpt-5.4"}`;
+}
+
+export function createMockQaProviderDefinition(
+  params: MockQaProviderDefinitionParams,
+): QaProviderDefinition {
+  return {
+    mode: params.mode,
+    kind: "mock",
+    standaloneCommand: {
+      name: params.commandName,
+      description: params.commandDescription,
+      serverLabel: params.serverLabel,
+    },
+    defaultModel: (options) => mockModelRef(params.mode, options?.alternate),
+    defaultImageGenerationProviderIds: [],
+    defaultImageGenerationModel: () => `${params.mode}/gpt-image-1`,
+    usesFastModeByDefault: () => false,
+    resolveModelParams: () => ({
+      transport: "sse",
+      openaiWsWarmup: false,
+    }),
+    resolveTurnTimeoutMs: ({ fallbackMs }) => fallbackMs,
+    buildGatewayModels: ({ providerBaseUrl }) => ({
+      mode: "replace",
+      providers: createMockProviderMap(params.mode, providerBaseUrl),
+    }),
+    mockAuthProviders: params.mockAuthProviders,
+    usesModelProviderPlugins: false,
+    scrubsLiveProviderEnv: true,
+    appliesLiveEnvAliases: false,
+  };
+}
--- a/extensions/qa-lab/src/providers/shared/types.ts
+++ b/extensions/qa-lab/src/providers/shared/types.ts
@@ -0,0 +1,56 @@
+import type { ModelProviderConfig } from "openclaw/plugin-sdk/provider-model-shared";
+import type { QaThinkingLevel } from "../../qa-thinking.js";
+
+export type QaProviderMode = "mock-openai" | "aimock" | "live-frontier";
+export type QaProviderModeInput = QaProviderMode;
+
+export type QaMockProviderServer = {
+  baseUrl: string;
+  stop(): Promise<void>;
+};
+
+export type QaProviderModelParamsInput = {
+  modelRef: string;
+  fastMode?: boolean;
+  thinkingDefault?: QaThinkingLevel;
+};
+
+export type QaProviderGatewayModelsInput = {
+  providerBaseUrl: string;
+  liveProviderConfigs?: Record<string, ModelProviderConfig>;
+};
+
+export type QaProviderDefaultImageInput = {
+  modelProviderIds: readonly string[];
+};
+
+export type QaProviderTurnTimeoutInput = {
+  primaryModel: string;
+  alternateModel: string;
+  modelRef: string;
+  fallbackMs: number;
+};
+
+export type QaProviderDefinition = {
+  mode: QaProviderMode;
+  kind: "mock" | "live";
+  standaloneCommand?: {
+    name: string;
+    description: string;
+    serverLabel: string;
+  };
+  defaultModel(options?: { alternate?: boolean; preferredLiveModel?: string }): string;
+  defaultImageGenerationProviderIds: readonly string[];
+  defaultImageGenerationModel(input: QaProviderDefaultImageInput): string | null;
+  usesFastModeByDefault(modelRef: string): boolean;
+  resolveModelParams(input: QaProviderModelParamsInput): Record<string, unknown>;
+  resolveTurnTimeoutMs(input: QaProviderTurnTimeoutInput): number;
+  buildGatewayModels(input: QaProviderGatewayModelsInput): {
+    mode: "replace" | "merge";
+    providers: Record<string, ModelProviderConfig>;
+  } | null;
+  mockAuthProviders?: readonly string[];
+  usesModelProviderPlugins: boolean;
+  scrubsLiveProviderEnv: boolean;
+  appliesLiveEnvAliases: boolean;
+};
--- a/extensions/qa-lab/src/qa-channel-transport.ts
+++ b/extensions/qa-lab/src/qa-channel-transport.ts
@@ -1,5 +1,6 @@
 import type { OpenClawConfig } from "openclaw/plugin-sdk/config-runtime";
 import type { QaBusState } from "./bus-state.js";
+import { getQaProvider } from "./providers/index.js";
 import { QaStateBackedTransportAdapter, waitForQaTransportCondition } from "./qa-transport.js";
 import type {
  QaTransportActionName,
@@ -71,9 +72,10 @@ export function createQaChannelGatewayConfig(params: {
 }

 function createQaChannelReportNotes(params: QaTransportReportParams) {
+  const provider = getQaProvider(params.providerMode);
  return [
-    params.providerMode === "mock-openai"
-      ? "Runs against qa-channel + qa-lab bus + real gateway child + mock OpenAI provider."
+    provider.kind === "mock"
+      ? `Runs against qa-channel + qa-lab bus + real gateway child + ${params.providerMode} provider.`
      : `Runs against qa-channel + qa-lab bus + real gateway child + live frontier models (${params.primaryModel}, ${params.alternateModel})${params.fastMode ? " with fast mode enabled" : ""}.`,
    params.concurrency > 1
      ? `Scenarios run in isolated gateway workers with concurrency ${params.concurrency}.`
--- a/extensions/qa-lab/src/qa-gateway-config.test.ts
+++ b/extensions/qa-lab/src/qa-gateway-config.test.ts
@@ -96,6 +96,29 @@ describe("buildQaGatewayConfig", () => {
    expect(cfg.plugins?.allow).toEqual(["memory-core"]);
  });

+  it("can wire AIMock as a separate mock provider lane", () => {
+    const cfg = buildQaGatewayConfig({
+      bind: "loopback",
+      gatewayPort: 18789,
+      gatewayToken: "token",
+      providerBaseUrl: "http://127.0.0.1:45080/v1",
+      workspaceDir: "/tmp/qa-workspace",
+      providerMode: "aimock",
+      primaryModel: "aimock/gpt-5.4",
+      alternateModel: "aimock/gpt-5.4-alt",
+    });
+
+    expect(getPrimaryModel(cfg.agents?.defaults?.model)).toBe("aimock/gpt-5.4");
+    expect(cfg.agents?.defaults?.imageGenerationModel).toEqual({
+      primary: "aimock/gpt-image-1",
+    });
+    expect(cfg.models?.providers?.aimock?.baseUrl).toBe("http://127.0.0.1:45080/v1");
+    expect(cfg.models?.providers?.aimock?.api).toBe("openai-responses");
+    expect(cfg.models?.providers?.openai?.baseUrl).toBe("http://127.0.0.1:45080/v1");
+    expect(cfg.models?.providers?.anthropic?.baseUrl).toBe("http://127.0.0.1:45080");
+    expect(cfg.models?.providers?.["mock-openai"]).toBeUndefined();
+  });
+
  it("can omit qa-channel for live transport gateway children", () => {
    const cfg = buildQaGatewayConfig({
      bind: "loopback",
--- a/extensions/qa-lab/src/qa-gateway-config.ts
+++ b/extensions/qa-lab/src/qa-gateway-config.ts
@@ -2,13 +2,17 @@ import type { OpenClawConfig } from "openclaw/plugin-sdk/config-runtime";
 import type { ModelProviderConfig } from "openclaw/plugin-sdk/provider-model-shared";
 import {
  defaultQaModelForMode,
-  isQaFastModeModelRef,
  normalizeQaProviderMode,
  splitQaModelRef,
  type QaProviderMode,
 } from "./model-selection.js";
+import { getQaProvider } from "./providers/index.js";
+import { DEFAULT_QA_PROVIDER_MODE } from "./providers/index.js";
+import { normalizeQaThinkingLevel, type QaThinkingLevel } from "./qa-thinking.js";
 import type { QaTransportGatewayConfig } from "./qa-transport.js";

+export { normalizeQaThinkingLevel, type QaThinkingLevel } from "./qa-thinking.js";
+
 export const DEFAULT_QA_CONTROL_UI_ALLOWED_ORIGINS = Object.freeze([
  "http://127.0.0.1:18789",
  "http://localhost:18789",
@@ -16,39 +20,6 @@ export const DEFAULT_QA_CONTROL_UI_ALLOWED_ORIGINS = Object.freeze([
  "http://localhost:43124",
 ]);

-export type QaThinkingLevel = "off" | "minimal" | "low" | "medium" | "high" | "xhigh" | "adaptive";
-
-export function normalizeQaThinkingLevel(input: unknown): QaThinkingLevel | undefined {
-  const value = typeof input === "string" ? input.trim().toLowerCase() : "";
-  const collapsed = value.replace(/[\s_-]+/g, "");
-  if (collapsed === "off") {
-    return "off";
-  }
-  if (collapsed === "minimal" || collapsed === "min") {
-    return "minimal";
-  }
-  if (collapsed === "low") {
-    return "low";
-  }
-  if (collapsed === "medium" || collapsed === "med") {
-    return "medium";
-  }
-  if (collapsed === "high") {
-    return "high";
-  }
-  if (collapsed === "xhigh" || collapsed === "extrahigh") {
-    return "xhigh";
-  }
-  if (collapsed === "adaptive" || collapsed === "auto") {
-    return "adaptive";
-  }
-  return undefined;
-}
-
-function trimTrailingApiV1(baseUrl: string) {
-  return baseUrl.replace(/\/v1\/?$/i, "");
-}
-
 export function mergeQaControlUiAllowedOrigins(extraOrigins?: string[]) {
  const normalizedExtra = (extraOrigins ?? [])
    .map((origin) => origin.trim())
@@ -65,7 +36,7 @@ export function buildQaGatewayConfig(params: {
  controlUiRoot?: string;
  controlUiAllowedOrigins?: string[];
  controlUiEnabled?: boolean;
-  providerMode?: QaProviderMode | "live-openai";
+  providerMode?: QaProviderMode;
  primaryModel?: string;
  alternateModel?: string;
  imageGenerationModel?: string | null;
@@ -77,108 +48,9 @@ export function buildQaGatewayConfig(params: {
  fastMode?: boolean;
  thinkingDefault?: QaThinkingLevel;
 }): OpenClawConfig {
-  const mockProviderBaseUrl = params.providerBaseUrl ?? "http://127.0.0.1:44080/v1";
-  const mockAnthropicBaseUrl = trimTrailingApiV1(mockProviderBaseUrl);
-  const mockOpenAiProvider: ModelProviderConfig = {
-    baseUrl: mockProviderBaseUrl,
-    apiKey: "test",
-    api: "openai-responses",
-    request: {
-      allowPrivateNetwork: true,
-    },
-    models: [
-      {
-        id: "gpt-5.4",
-        name: "gpt-5.4",
-        api: "openai-responses",
-        reasoning: false,
-        input: ["text", "image"],
-        cost: {
-          input: 0,
-          output: 0,
-          cacheRead: 0,
-          cacheWrite: 0,
-        },
-        contextWindow: 128_000,
-        maxTokens: 4096,
-      },
-      {
-        id: "gpt-5.4-alt",
-        name: "gpt-5.4-alt",
-        api: "openai-responses",
-        reasoning: false,
-        input: ["text", "image"],
-        cost: {
-          input: 0,
-          output: 0,
-          cacheRead: 0,
-          cacheWrite: 0,
-        },
-        contextWindow: 128_000,
-        maxTokens: 4096,
-      },
-      {
-        id: "gpt-image-1",
-        name: "gpt-image-1",
-        api: "openai-responses",
-        reasoning: false,
-        input: ["text"],
-        cost: {
-          input: 0,
-          output: 0,
-          cacheRead: 0,
-          cacheWrite: 0,
-        },
-        contextWindow: 128_000,
-        maxTokens: 4096,
-      },
-    ],
-  };
-  const mockNamedOpenAiProvider: ModelProviderConfig = {
-    ...mockOpenAiProvider,
-    models: mockOpenAiProvider.models.map((model) => ({ ...model })),
-  };
-  const mockAnthropicProvider: ModelProviderConfig = {
-    baseUrl: mockAnthropicBaseUrl,
-    apiKey: "test",
-    api: "anthropic-messages",
-    request: {
-      allowPrivateNetwork: true,
-    },
-    models: [
-      {
-        id: "claude-opus-4-6",
-        name: "claude-opus-4-6",
-        api: "anthropic-messages",
-        reasoning: false,
-        input: ["text", "image"],
-        cost: {
-          input: 0,
-          output: 0,
-          cacheRead: 0,
-          cacheWrite: 0,
-        },
-        contextWindow: 200_000,
-        maxTokens: 4096,
-      },
-      {
-        id: "claude-sonnet-4-6",
-        name: "claude-sonnet-4-6",
-        api: "anthropic-messages",
-        reasoning: false,
-        input: ["text", "image"],
-        cost: {
-          input: 0,
-          output: 0,
-          cacheRead: 0,
-          cacheWrite: 0,
-        },
-        contextWindow: 200_000,
-        maxTokens: 4096,
-      },
-    ],
-  };
-  const providerMode = normalizeQaProviderMode(params.providerMode ?? "mock-openai");
+  const providerBaseUrl = params.providerBaseUrl ?? "http://127.0.0.1:44080/v1";
+  const providerMode = normalizeQaProviderMode(params.providerMode ?? DEFAULT_QA_PROVIDER_MODE);
+  const provider = getQaProvider(providerMode);
  const primaryModel = params.primaryModel ?? defaultQaModelForMode(providerMode);
  const alternateModel =
    params.alternateModel ?? defaultQaModelForMode(providerMode, { alternate: true });
@@ -188,39 +60,33 @@ export function buildQaGatewayConfig(params: {
  const imageGenerationModelRef =
    params.imageGenerationModel !== undefined
      ? params.imageGenerationModel
-      : providerMode === "mock-openai"
-        ? "mock-openai/gpt-image-1"
-        : modelProviderIds.includes("openai")
-          ? "openai/gpt-image-1"
-          : null;
-  const selectedProviderIds =
-    providerMode === "live-frontier"
-      ? [
-          ...new Set(
-            [...(params.enabledProviderIds ?? []), ...modelProviderIds, imageGenerationModelRef]
-              .map((value) =>
-                typeof value === "string" ? (splitQaModelRef(value)?.provider ?? value) : null,
-              )
-              .filter((provider): provider is string => Boolean(provider)),
-          ),
-        ]
-      : [];
-  const selectedPluginIds =
-    providerMode === "live-frontier"
-      ? [
-          ...new Set(
-            (params.enabledPluginIds?.length ?? 0) > 0
-              ? params.enabledPluginIds
-              : selectedProviderIds,
-          ),
-        ]
-      : [
-          ...new Set(
-            (params.enabledPluginIds ?? [])
-              .map((pluginId) => pluginId.trim())
-              .filter((pluginId) => pluginId.length > 0),
-          ),
-        ];
+      : provider.defaultImageGenerationModel({ modelProviderIds });
+  const selectedProviderIds = provider.usesModelProviderPlugins
+    ? [
+        ...new Set(
+          [...(params.enabledProviderIds ?? []), ...modelProviderIds, imageGenerationModelRef]
+            .map((value) =>
+              typeof value === "string" ? (splitQaModelRef(value)?.provider ?? value) : null,
+            )
+            .filter((provider): provider is string => Boolean(provider)),
+        ),
+      ]
+    : [];
+  const selectedPluginIds = provider.usesModelProviderPlugins
+    ? [
+        ...new Set(
+          (params.enabledPluginIds?.length ?? 0) > 0
+            ? params.enabledPluginIds
+            : selectedProviderIds,
+        ),
+      ]
+    : [
+        ...new Set(
+          (params.enabledPluginIds ?? [])
+            .map((pluginId) => pluginId.trim())
+            .filter((pluginId) => pluginId.length > 0),
+        ),
+      ];
  const transportPluginIds = [...new Set(params.transportPluginIds ?? [])]
    .map((pluginId) => pluginId.trim())
    .filter((pluginId) => pluginId.length > 0);
@@ -231,22 +97,17 @@ export function buildQaGatewayConfig(params: {
    transportPluginIds.map((pluginId) => [pluginId, { enabled: true }]),
  );
  const allowedPlugins = [...new Set(["memory-core", ...selectedPluginIds, ...transportPluginIds])];
-  const liveModelParams =
-    providerMode === "live-frontier"
-      ? (modelRef: string) => ({
-          transport: "sse",
-          openaiWsWarmup: false,
-          ...(params.fastMode === true || isQaFastModeModelRef(modelRef) ? { fastMode: true } : {}),
-          ...(params.thinkingDefault ? { thinking: params.thinkingDefault } : {}),
-        })
-      : (_modelRef: string) => ({
-          transport: "sse",
-          openaiWsWarmup: false,
-        });
+  const resolveModelParams = (modelRef: string) =>
+    provider.resolveModelParams({
+      modelRef,
+      fastMode: params.fastMode,
+      thinkingDefault: params.thinkingDefault,
+    });
  const allowedOrigins = mergeQaControlUiAllowedOrigins(params.controlUiAllowedOrigins);
-  const liveProviderConfigs =
-    providerMode === "live-frontier" ? (params.liveProviderConfigs ?? {}) : {};
-  const hasLiveProviderConfigs = Object.keys(liveProviderConfigs).length > 0;
+  const gatewayModels = provider.buildGatewayModels({
+    providerBaseUrl,
+    liveProviderConfigs: params.liveProviderConfigs,
+  });

  return {
    plugins: {
@@ -286,10 +147,10 @@ export function buildQaGatewayConfig(params: {
        },
        models: {
          [primaryModel]: {
-            params: liveModelParams(primaryModel),
+            params: resolveModelParams(primaryModel),
          },
          [alternateModel]: {
-            params: liveModelParams(alternateModel),
+            params: resolveModelParams(alternateModel),
          },
        },
        subagents: {
@@ -319,25 +180,14 @@ export function buildQaGatewayConfig(params: {
    memory: {
      backend: "builtin",
    },
-    ...(providerMode === "mock-openai"
+    ...(gatewayModels
      ? {
          models: {
-            mode: "replace",
-            providers: {
-              "mock-openai": mockOpenAiProvider,
-              openai: mockNamedOpenAiProvider,
-              anthropic: mockAnthropicProvider,
-            },
+            mode: gatewayModels.mode,
+            providers: gatewayModels.providers,
          },
        }
-      : hasLiveProviderConfigs
-        ? {
-            models: {
-              mode: "merge",
-              providers: liveProviderConfigs,
-            },
-          }
-        : {}),
+      : {}),
    gateway: {
      mode: "local",
      bind: params.bind,
--- a/extensions/qa-lab/src/qa-thinking.ts
+++ b/extensions/qa-lab/src/qa-thinking.ts
@@ -0,0 +1,28 @@
+export type QaThinkingLevel = "off" | "minimal" | "low" | "medium" | "high" | "xhigh" | "adaptive";
+
+export function normalizeQaThinkingLevel(input: unknown): QaThinkingLevel | undefined {
+  const value = typeof input === "string" ? input.trim().toLowerCase() : "";
+  const collapsed = value.replace(/[\s_-]+/g, "");
+  if (collapsed === "off") {
+    return "off";
+  }
+  if (collapsed === "minimal" || collapsed === "min") {
+    return "minimal";
+  }
+  if (collapsed === "low") {
+    return "low";
+  }
+  if (collapsed === "medium" || collapsed === "med") {
+    return "medium";
+  }
+  if (collapsed === "high") {
+    return "high";
+  }
+  if (collapsed === "xhigh" || collapsed === "extrahigh") {
+    return "xhigh";
+  }
+  if (collapsed === "adaptive" || collapsed === "auto") {
+    return "adaptive";
+  }
+  return undefined;
+}
--- a/extensions/qa-lab/src/qa-transport.ts
+++ b/extensions/qa-lab/src/qa-transport.ts
@@ -1,5 +1,6 @@
 import { setTimeout as sleep } from "node:timers/promises";
 import type { OpenClawConfig } from "openclaw/plugin-sdk/config-runtime";
+import type { QaProviderMode } from "./model-selection.js";
 import { extractQaFailureReplyText } from "./reply-failure.js";
 import type {
  QaBusInboundMessageInput,
@@ -24,7 +25,7 @@ export type QaTransportGatewayClient = {
 export type QaTransportActionName = "delete" | "edit" | "react" | "thread-create";

 export type QaTransportReportParams = {
-  providerMode: "mock-openai" | "live-frontier";
+  providerMode: QaProviderMode;
  primaryModel: string;
  alternateModel: string;
  fastMode: boolean;
--- a/extensions/qa-lab/src/run-config.test.ts
+++ b/extensions/qa-lab/src/run-config.test.ts
@@ -1,24 +1,20 @@
 import { beforeEach, describe, expect, it, vi } from "vitest";

 const { defaultQaRuntimeModelForMode } = vi.hoisted(() => ({
-  defaultQaRuntimeModelForMode: vi.fn<(mode: string, options?: { alternate?: boolean }) => string>(
-    (mode, options) =>
-      mode === "live-frontier"
-        ? "openai/gpt-5.4"
-        : options?.alternate
-          ? "mock-openai/gpt-5.4-alt"
-          : "mock-openai/gpt-5.4",
-  ),
+  defaultQaRuntimeModelForMode:
+    vi.fn<(mode: string, options?: { alternate?: boolean }) => string>(),
 }));

 vi.mock("./model-selection.runtime.js", () => ({
  defaultQaRuntimeModelForMode,
 }));
+import { defaultQaModelForMode as defaultQaProviderModelForMode } from "./model-selection.js";
 import {
  createDefaultQaRunSelection,
  createIdleQaRunnerSnapshot,
  createQaRunOutputDir,
  normalizeQaRunSelection,
+  type QaProviderModeInput,
 } from "./run-config.js";

 const scenarios = [
@@ -42,11 +38,7 @@ describe("qa run config", () => {
  beforeEach(() => {
    defaultQaRuntimeModelForMode.mockImplementation(
      (mode: string, options?: { alternate?: boolean }) =>
-        mode === "live-frontier"
-          ? "openai/gpt-5.4"
-          : options?.alternate
-            ? "mock-openai/gpt-5.4-alt"
-            : "mock-openai/gpt-5.4",
+        defaultQaProviderModelForMode(mode as QaProviderModeInput, options),
    );
  });

@@ -64,7 +56,7 @@ describe("qa run config", () => {
    expect(
      normalizeQaRunSelection(
        {
-          providerMode: "live-openai",
+          providerMode: "live-frontier",
          primaryModel: "openai/gpt-5.4",
          alternateModel: "",
          fastMode: false,
@@ -81,6 +73,17 @@ describe("qa run config", () => {
    });
  });

+  it("rejects removed provider compatibility names", () => {
+    expect(() =>
+      normalizeQaRunSelection(
+        {
+          providerMode: "live-openai",
+        },
+        scenarios,
+      ),
+    ).toThrow("unknown QA provider mode: live-openai");
+  });
+
  it("falls back to all scenarios when selection would otherwise be empty", () => {
    const snapshot = createIdleQaRunnerSnapshot(scenarios);
    expect(snapshot.status).toBe("idle");
@@ -95,6 +98,26 @@ describe("qa run config", () => {
    ).toEqual(["dm-chat-baseline", "thread-lifecycle"]);
  });

+  it("normalizes aimock selections", () => {
+    expect(
+      normalizeQaRunSelection(
+        {
+          providerMode: "aimock",
+          primaryModel: "",
+          alternateModel: "",
+          scenarioIds: ["dm-chat-baseline"],
+        },
+        scenarios,
+      ),
+    ).toEqual({
+      providerMode: "aimock",
+      primaryModel: "aimock/gpt-5.4",
+      alternateModel: "aimock/gpt-5.4-alt",
+      fastMode: false,
+      scenarioIds: ["dm-chat-baseline"],
+    });
+  });
+
  it("anchors generated run output dirs under the provided repo root", () => {
    const outputDir = createQaRunOutputDir("/tmp/openclaw-repo");
    expect(outputDir.startsWith("/tmp/openclaw-repo/.artifacts/qa-e2e/lab-")).toBe(true);
@@ -104,9 +127,7 @@ describe("qa run config", () => {
    defaultQaRuntimeModelForMode.mockImplementation((mode, options) =>
      mode === "live-frontier"
        ? "openai-codex/gpt-5.4"
-        : options?.alternate
-          ? "mock-openai/gpt-5.4-alt"
-          : "mock-openai/gpt-5.4",
+        : defaultQaProviderModelForMode(mode as QaProviderModeInput, options),
    );

    expect(createDefaultQaRunSelection(scenarios)).toEqual({
--- a/extensions/qa-lab/src/run-config.ts
+++ b/extensions/qa-lab/src/run-config.ts
@@ -1,13 +1,16 @@
 import path from "node:path";
+import { defaultQaRuntimeModelForMode } from "./model-selection.runtime.js";
 import {
+  DEFAULT_QA_LIVE_PROVIDER_MODE,
+  getQaProvider,
+  isQaProviderModeInput,
  normalizeQaProviderMode as normalizeQaProviderModeInput,
  type QaProviderMode,
-} from "./model-selection.js";
-import { defaultQaRuntimeModelForMode } from "./model-selection.runtime.js";
+} from "./providers/index.js";
 import type { QaSeedScenario } from "./scenario-catalog.js";

 export type { QaProviderMode } from "./model-selection.js";
-export type QaProviderModeInput = QaProviderMode | "live-openai";
+export type { QaProviderModeInput } from "./providers/index.js";

 export type QaLabRunSelection = {
  providerMode: QaProviderMode;
@@ -38,7 +41,7 @@ export function defaultQaModelForMode(mode: QaProviderMode, alternate = false) {
 }

 export function createDefaultQaRunSelection(scenarios: QaSeedScenario[]): QaLabRunSelection {
-  const providerMode: QaProviderMode = "live-frontier";
+  const providerMode: QaProviderMode = DEFAULT_QA_LIVE_PROVIDER_MODE;
  return {
    providerMode,
    primaryModel: defaultQaModelForMode(providerMode),
@@ -49,11 +52,14 @@ export function createDefaultQaRunSelection(scenarios: QaSeedScenario[]): QaLabR
 }

 export function normalizeQaProviderMode(input: unknown): QaProviderMode {
-  return normalizeQaProviderModeInput(
-    input === "mock-openai" || input === "live-frontier" || input === "live-openai"
-      ? input
-      : "live-frontier",
-  );
+  if (input === undefined || input === null || input === "") {
+    return DEFAULT_QA_LIVE_PROVIDER_MODE;
+  }
+  if (isQaProviderModeInput(input)) {
+    return normalizeQaProviderModeInput(input);
+  }
+  const details = typeof input === "string" ? `: ${input}` : "";
+  throw new Error(`unknown QA provider mode${details}`);
 }

 function normalizeModel(input: unknown, fallback: string) {
@@ -87,7 +93,7 @@ export function normalizeQaRunSelection(
      payload.alternateModel,
      defaultQaModelForMode(providerMode, true),
    ),
-    fastMode: providerMode === "live-frontier" || payload.fastMode === true,
+    fastMode: getQaProvider(providerMode).kind === "live" || payload.fastMode === true,
    scenarioIds: normalizeScenarioIds(payload.scenarioIds, scenarios),
  };
 }
--- a/extensions/qa-lab/src/suite-planning.ts
+++ b/extensions/qa-lab/src/suite-planning.ts
@@ -2,6 +2,8 @@ import path from "node:path";
 import { normalizeLowercaseStringOrEmpty } from "openclaw/plugin-sdk/text-runtime";
 import { ensureRepoBoundDirectory, resolveRepoRelativeOutputDir } from "./cli-paths.js";
 import type { QaCliBackendAuthMode } from "./gateway-child.js";
+import type { QaProviderMode } from "./model-selection.js";
+import { getQaProvider } from "./providers/index.js";
 import type { QaTransportId } from "./qa-transport-registry.js";
 import { readQaBootstrapScenarioCatalog } from "./scenario-catalog.js";

@@ -28,10 +30,10 @@ function normalizeQaConfigString(value: unknown): string | undefined {
 function scenarioMatchesLiveLane(params: {
  scenario: QaSeedScenario;
  primaryModel: string;
-  providerMode: "mock-openai" | "live-frontier";
+  providerMode: QaProviderMode;
  claudeCliAuthMode?: QaCliBackendAuthMode;
 }) {
-  if (params.providerMode !== "live-frontier") {
+  if (getQaProvider(params.providerMode).kind !== "live") {
    return true;
  }
  const selected = splitModelRef(params.primaryModel);
@@ -54,7 +56,7 @@ function scenarioMatchesLiveLane(params: {
 function selectQaSuiteScenarios(params: {
  scenarios: ReturnType<typeof readQaBootstrapScenarioCatalog>["scenarios"];
  scenarioIds?: string[];
-  providerMode: "mock-openai" | "live-frontier";
+  providerMode: QaProviderMode;
  primaryModel: string;
  claudeCliAuthMode?: QaCliBackendAuthMode;
 }) {
--- a/extensions/qa-lab/src/suite-runtime-agent-common.ts
+++ b/extensions/qa-lab/src/suite-runtime-agent-common.ts
@@ -1,7 +1,8 @@
 import { resolveQaLiveTurnTimeoutMs } from "./live-timeout.js";
+import type { QaProviderMode } from "./model-selection.js";

 type QaLiveTimeoutEnv = {
-  providerMode: "mock-openai" | "live-frontier";
+  providerMode: QaProviderMode;
  primaryModel: string;
  alternateModel: string;
 };
--- a/extensions/qa-lab/src/suite-runtime-agent-media.test.ts
+++ b/extensions/qa-lab/src/suite-runtime-agent-media.test.ts
@@ -91,7 +91,7 @@ describe("qa suite runtime agent media helpers", () => {
    ).resolves.toBe(mediaPath);
  });

-  it("applies mock image generation config with transport-required plugins", async () => {
+  it("applies provider image generation config with transport-required plugins", async () => {
    await ensureImageGenerationConfigured({
      providerMode: "mock-openai",
      mock: { baseUrl: "http://127.0.0.1:9999" },
@@ -102,7 +102,7 @@ describe("qa suite runtime agent media helpers", () => {
      expect.objectContaining({
        patch: expect.objectContaining({
          plugins: expect.objectContaining({
-            allow: expect.arrayContaining(["memory-core", "openai", "qa-channel", "browser"]),
+            allow: expect.arrayContaining(["memory-core", "qa-channel", "browser"]),
          }),
        }),
      }),
--- a/extensions/qa-lab/src/suite-runtime-agent-media.ts
+++ b/extensions/qa-lab/src/suite-runtime-agent-media.ts
@@ -1,5 +1,6 @@
 import fs from "node:fs/promises";
 import path from "node:path";
+import { buildQaImageGenerationConfigPatch } from "./providers/image-generation.js";
 import {
  fetchJson,
  patchConfig,
@@ -70,63 +71,13 @@ async function resolveGeneratedImagePath(params: {
 }

 async function ensureImageGenerationConfigured(env: QaSuiteRuntimeEnv) {
-  const imageModelRef = "openai/gpt-image-1";
  await patchConfig({
    env,
-    patch:
-      env.providerMode === "mock-openai"
-        ? {
-            plugins: {
-              allow: [...new Set(["memory-core", "openai", ...env.transport.requiredPluginIds])],
-              entries: {
-                openai: {
-                  enabled: true,
-                },
-              },
-            },
-            models: {
-              providers: {
-                openai: {
-                  baseUrl: `${env.mock?.baseUrl}/v1`,
-                  apiKey: "test",
-                  api: "openai-responses",
-                  models: [
-                    {
-                      id: "gpt-image-1",
-                      name: "gpt-image-1",
-                      api: "openai-responses",
-                      reasoning: false,
-                      input: ["text"],
-                      cost: {
-                        input: 0,
-                        output: 0,
-                        cacheRead: 0,
-                        cacheWrite: 0,
-                      },
-                      contextWindow: 128_000,
-                      maxTokens: 4096,
-                    },
-                  ],
-                },
-              },
-            },
-            agents: {
-              defaults: {
-                imageGenerationModel: {
-                  primary: imageModelRef,
-                },
-              },
-            },
-          }
-        : {
-            agents: {
-              defaults: {
-                imageGenerationModel: {
-                  primary: imageModelRef,
-                },
-              },
-            },
-          },
+    patch: buildQaImageGenerationConfigPatch({
+      providerMode: env.providerMode,
+      providerBaseUrl: env.mock ? `${env.mock.baseUrl}/v1` : undefined,
+      requiredPluginIds: env.transport.requiredPluginIds,
+    }),
  });
  await waitForGatewayHealthy(env);
  await waitForTransportReady(env, 60_000);
--- a/extensions/qa-lab/src/suite-runtime-types.ts
+++ b/extensions/qa-lab/src/suite-runtime-types.ts
@@ -1,4 +1,5 @@
 import type { OpenClawConfig } from "openclaw/plugin-sdk/config-runtime";
+import type { QaProviderMode } from "./model-selection.js";
 import type { QaTransportActionName, QaTransportAdapter } from "./qa-transport.js";

 export type QaRuntimeGatewayClient = {
@@ -21,7 +22,7 @@ export type QaSuiteRuntimeEnv = {
  gateway: QaRuntimeGatewayClient;
  transport: QaRuntimeTransport;
  repoRoot: string;
-  providerMode: "mock-openai" | "live-frontier";
+  providerMode: QaProviderMode;
  primaryModel: string;
  alternateModel: string;
  mock: {
--- a/extensions/qa-lab/src/suite.ts
+++ b/extensions/qa-lab/src/suite.ts
@@ -12,12 +12,13 @@ import type {
  QaLabServerStartParams,
 } from "./lab-server.types.js";
 import { resolveQaLiveTurnTimeoutMs } from "./live-timeout.js";
-import { startQaMockOpenAiServer } from "./mock-openai-server.js";
 import {
  isQaFastModeEnabled,
  normalizeQaProviderMode,
  type QaProviderMode,
 } from "./model-selection.js";
+import { DEFAULT_QA_LIVE_PROVIDER_MODE } from "./providers/index.js";
+import { startQaProviderServer } from "./providers/server-runtime.js";
 import type { QaThinkingLevel } from "./qa-gateway-config.js";
 import {
  createQaTransportAdapter,
@@ -69,7 +70,7 @@ export type QaSuiteStartLabFn = (params?: QaLabServerStartParams) => Promise<QaL
 export type QaSuiteRunParams = {
  repoRoot?: string;
  outputDir?: string;
-  providerMode?: QaProviderMode | "live-openai";
+  providerMode?: QaProviderMode;
  transportId?: QaTransportId;
  primaryModel?: string;
  alternateModel?: string;
@@ -195,7 +196,7 @@ async function runScenarioDefinition(

 function createQaSuiteReportNotes(params: {
  transport: QaTransportAdapter;
-  providerMode: "mock-openai" | "live-frontier";
+  providerMode: QaProviderMode;
  primaryModel: string;
  alternateModel: string;
  fastMode: boolean;
@@ -331,7 +332,9 @@ async function writeQaSuiteArtifacts(params: {
 export async function runQaSuite(params?: QaSuiteRunParams): Promise<QaSuiteResult> {
  const startedAt = new Date();
  const repoRoot = path.resolve(params?.repoRoot ?? process.cwd());
-  const providerMode = normalizeQaProviderMode(params?.providerMode ?? "live-frontier");
+  const providerMode = normalizeQaProviderMode(
+    params?.providerMode ?? DEFAULT_QA_LIVE_PROVIDER_MODE,
+  );
  const transportId = normalizeQaTransportId(params?.transportId);
  const primaryModel = params?.primaryModel ?? defaultQaModelForMode(providerMode);
  const alternateModel = params?.alternateModel ?? defaultQaModelForMode(providerMode, true);
@@ -538,13 +541,7 @@ export async function runQaSuite(params?: QaSuiteRunParams): Promise<QaSuiteResu
    id: transportId,
    state: lab.state,
  });
-  const mock =
-    providerMode === "mock-openai"
-      ? await startQaMockOpenAiServer({
-          host: "127.0.0.1",
-          port: 0,
-        })
-      : null;
+  const mock = await startQaProviderServer(providerMode);
  const gateway = await startQaGatewayChild({
    repoRoot,
    providerBaseUrl: mock ? `${mock.baseUrl}/v1` : undefined,
--- a/package.json
+++ b/package.json
@@ -1451,6 +1451,7 @@
    "zod": "^4.3.6"
  },
  "devDependencies": {
+    "@copilotkit/aimock": "1.13.0",
    "@grammyjs/types": "^3.26.0",
    "@lit-labs/signals": "^0.2.0",
    "@lit/context": "^1.1.6",
--- a/pnpm-lock.yaml
+++ b/pnpm-lock.yaml
@@ -255,6 +255,9 @@ importers:
        specifier: ^4.3.6
        version: 4.3.6
    devDependencies:
+      '@copilotkit/aimock':
+        specifier: 1.13.0
+        version: 1.13.0
      '@grammyjs/types':
        specifier: ^3.26.0
        version: 3.26.0
@@ -1734,6 +1737,11 @@ packages:
    resolution: {integrity: sha512-ooWCrlZP11i8GImSjTHYHLkvFDP48nS4+204nGb1RiX/WXYHmJA2III9/e2DWVabCESdW7hBAEzHRqUn9OUVvQ==}
    engines: {node: '>=0.1.90'}

+  '@copilotkit/aimock@1.13.0':
+    resolution: {integrity: sha512-wcNQdicxSuaHf4MLtoC9bouKeYGcV9aBoIVJbeLtrBUhqz5LUAmo/U/S1WT69kyAEuM0uO1XyD3Jz+V+H9jsUw==}
+    engines: {node: '>=20.15.0'}
+    hasBin: true
+
  '@create-markdown/preview@2.0.3':
    resolution: {integrity: sha512-Vrp8DyuiouryZ3E4NQ7tBgoYQdoekd0+DzN64mZ48QYCw3V+MCb/H2q10SW8KC8XPr931XOMDvKX4I83qpQh3g==}
    engines: {node: '>=20.0.0'}
@@ -8749,6 +8757,8 @@ snapshots:
  '@colors/colors@1.5.0':
    optional: true

+  '@copilotkit/aimock@1.13.0': {}
+
  '@create-markdown/preview@2.0.3(shiki@3.23.0)':
    optionalDependencies:
      shiki: 3.23.0