refactor(providers): remove core default and usage bias

docs/automation/cron-jobs.md

@@ -163,7 +163,7 @@ Run an isolated agent turn:
 curl -X POST http://127.0.0.1:18789/hooks/agent \
   -H 'Authorization: Bearer SECRET' \
   -H 'Content-Type: application/json' \
-  -d '{"message":"Summarize inbox","name":"Email","model":"openai/gpt-5.2-mini"}'
+  -d '{"message":"Summarize inbox","name":"Email","model":"openai/gpt-5.4-mini"}'
 ```
 
 Fields: `message` (required), `name`, `agentId`, `wakeMode`, `deliver`, `channel`, `to`, `model`, `thinking`, `timeoutSeconds`.

docs/concepts/models.md

@@ -99,7 +99,7 @@ You can switch models for the current session without restarting:
 /model
 /model list
 /model 3
-/model openai/gpt-5.2
+/model openai/gpt-5.4
 /model status
 ```
 

docs/gateway/configuration-reference.md

@@ -980,7 +980,7 @@ Time format in system prompt. Default: `auto` (OS preference).
 - `pdfMaxPages`: default maximum pages considered by extraction fallback mode in the `pdf` tool.
 - `verboseDefault`: default verbose level for agents. Values: `"off"`, `"on"`, `"full"`. Default: `"off"`.
 - `elevatedDefault`: default elevated-output level for agents. Values: `"off"`, `"on"`, `"ask"`, `"full"`. Default: `"on"`.
-- `model.primary`: format `provider/model` (e.g. `anthropic/claude-opus-4-6`). If you omit the provider, OpenClaw assumes `anthropic` (deprecated).
+- `model.primary`: format `provider/model` (e.g. `openai/gpt-5.4`). If you omit the provider, OpenClaw assumes the configured default provider (currently `openai`; deprecated fallback behavior, so prefer explicit `provider/model`).
 - `models`: the configured model catalog and allowlist for `/model`. Each entry can include `alias` (shortcut) and `params` (provider-specific, for example `temperature`, `maxTokens`, `cacheRetention`, `context1m`).
 - `params`: global default provider parameters applied to all models. Set at `agents.defaults.params` (e.g. `{ cacheRetention: "long" }`).
 - `params` merge precedence (config): `agents.defaults.params` (global base) is overridden by `agents.defaults.models["provider/model"].params` (per-model), then `agents.list[].params` (matching agent id) overrides by key. See [Prompt Caching](/reference/prompt-caching) for details.

docs/gateway/configuration.md

@@ -113,11 +113,11 @@ When validation fails:
         defaults: {
           model: {
             primary: "anthropic/claude-sonnet-4-6",
-            fallbacks: ["openai/gpt-5.2"],
+            fallbacks: ["openai/gpt-5.4"],
           },
           models: {
             "anthropic/claude-sonnet-4-6": { alias: "Sonnet" },
-            "openai/gpt-5.2": { alias: "GPT" },
+            "openai/gpt-5.4": { alias: "GPT" },
           },
         },
       },

docs/help/faq.md

@@ -2057,7 +2057,7 @@ for usage/billing and raise limits as needed.
     agents.defaults.model.primary
     ```
 
-    Models are referenced as `provider/model` (example: `anthropic/claude-opus-4-6`). If you omit the provider, OpenClaw currently assumes `anthropic` as a temporary deprecation fallback - but you should still **explicitly** set `provider/model`.
+    Models are referenced as `provider/model` (example: `openai/gpt-5.4`). If you omit the provider, OpenClaw currently assumes the configured default provider (currently `openai`) as a temporary deprecation fallback - but you should still **explicitly** set `provider/model`.
 
   </Accordion>
 

docs/help/testing.md

@@ -193,7 +193,7 @@ Live tests are split into two layers so we can isolate failures:
 - How to select models:
   - `OPENCLAW_LIVE_MODELS=modern` to run the modern allowlist (Opus/Sonnet 4.6+, GPT-5.x + Codex, Gemini 3, GLM 4.7, MiniMax M2.7, Grok 4)
   - `OPENCLAW_LIVE_MODELS=all` is an alias for the modern allowlist
-  - or `OPENCLAW_LIVE_MODELS="openai/gpt-5.2,anthropic/claude-opus-4-6,..."` (comma allowlist)
+  - or `OPENCLAW_LIVE_MODELS="openai/gpt-5.4,anthropic/claude-opus-4-6,..."` (comma allowlist)
 - How to select providers:
   - `OPENCLAW_LIVE_PROVIDERS="google,google-antigravity,google-gemini-cli"` (comma allowlist)
 - Where keys come from:
@@ -356,13 +356,13 @@ Docker notes:
 Narrow, explicit allowlists are fastest and least flaky:
 
 - Single model, direct (no gateway):
-  - `OPENCLAW_LIVE_MODELS="openai/gpt-5.2" pnpm test:live src/agents/models.profiles.live.test.ts`
+  - `OPENCLAW_LIVE_MODELS="openai/gpt-5.4" pnpm test:live src/agents/models.profiles.live.test.ts`
 
 - Single model, gateway smoke:
-  - `OPENCLAW_LIVE_GATEWAY_MODELS="openai/gpt-5.2" pnpm test:live src/gateway/gateway-models.profiles.live.test.ts`
+  - `OPENCLAW_LIVE_GATEWAY_MODELS="openai/gpt-5.4" pnpm test:live src/gateway/gateway-models.profiles.live.test.ts`
 
 - Tool calling across several providers:
-  - `OPENCLAW_LIVE_GATEWAY_MODELS="openai/gpt-5.2,anthropic/claude-opus-4-6,google/gemini-3-flash-preview,zai/glm-4.7,minimax/MiniMax-M2.7" pnpm test:live src/gateway/gateway-models.profiles.live.test.ts`
+  - `OPENCLAW_LIVE_GATEWAY_MODELS="openai/gpt-5.4,anthropic/claude-opus-4-6,google/gemini-3-flash-preview,zai/glm-4.7,minimax/MiniMax-M2.7" pnpm test:live src/gateway/gateway-models.profiles.live.test.ts`
 
 - Google focus (Gemini API key + Antigravity):
   - Gemini (API key): `OPENCLAW_LIVE_GATEWAY_MODELS="google/gemini-3-flash-preview" pnpm test:live src/gateway/gateway-models.profiles.live.test.ts`
@@ -385,7 +385,7 @@ There is no fixed “CI model list” (live is opt-in), but these are the **reco
 
 This is the “common models” run we expect to keep working:
 
-- OpenAI (non-Codex): `openai/gpt-5.2` (optional: `openai/gpt-5.1`)
+- OpenAI (non-Codex): `openai/gpt-5.4` (optional: `openai/gpt-5.4-mini`)
 - OpenAI Codex: `openai-codex/gpt-5.4`
 - Anthropic: `anthropic/claude-opus-4-6` (or `anthropic/claude-sonnet-4-6`)
 - Google (Gemini API): `google/gemini-3.1-pro-preview` and `google/gemini-3-flash-preview` (avoid older Gemini 2.x models)
@@ -394,13 +394,13 @@ This is the “common models” run we expect to keep working:
 - MiniMax: `minimax/MiniMax-M2.7`
 
 Run gateway smoke with tools + image:
-`OPENCLAW_LIVE_GATEWAY_MODELS="openai/gpt-5.2,openai-codex/gpt-5.4,anthropic/claude-opus-4-6,google/gemini-3.1-pro-preview,google/gemini-3-flash-preview,google-antigravity/claude-opus-4-6-thinking,google-antigravity/gemini-3-flash,zai/glm-4.7,minimax/MiniMax-M2.7" pnpm test:live src/gateway/gateway-models.profiles.live.test.ts`
+`OPENCLAW_LIVE_GATEWAY_MODELS="openai/gpt-5.4,openai-codex/gpt-5.4,anthropic/claude-opus-4-6,google/gemini-3.1-pro-preview,google/gemini-3-flash-preview,google-antigravity/claude-opus-4-6-thinking,google-antigravity/gemini-3-flash,zai/glm-4.7,minimax/MiniMax-M2.7" pnpm test:live src/gateway/gateway-models.profiles.live.test.ts`
 
 ### Baseline: tool calling (Read + optional Exec)
 
 Pick at least one per provider family:
 
-- OpenAI: `openai/gpt-5.2` (or `openai/gpt-5-mini`)
+- OpenAI: `openai/gpt-5.4` (or `openai/gpt-5.4-mini`)
 - Anthropic: `anthropic/claude-opus-4-6` (or `anthropic/claude-sonnet-4-6`)
 - Google: `google/gemini-3-flash-preview` (or `google/gemini-3.1-pro-preview`)
 - Z.AI (GLM): `zai/glm-4.7`

src/agents/defaults.ts

@@ -1,6 +1,6 @@
 // Defaults for agent metadata when upstream does not supply them.
-// Model id uses pi-ai's built-in Anthropic catalog.
-export const DEFAULT_PROVIDER = "anthropic";
-export const DEFAULT_MODEL = "claude-opus-4-6";
+// Keep this aligned with the product-level latest-model baseline.
+export const DEFAULT_PROVIDER = "openai";
+export const DEFAULT_MODEL = "gpt-5.4";
 // Conservative fallback used when model metadata is unavailable.
 export const DEFAULT_CONTEXT_TOKENS = 200_000;

src/agents/model-selection.test.ts

@@ -101,8 +101,8 @@ function createProviderWithModelsConfig(provider: string, models: Array<Record<s
 function resolveConfiguredRefForTest(cfg: Partial<OpenClawConfig>) {
   return resolveConfiguredModelRef({
     cfg: cfg as OpenClawConfig,
-    defaultProvider: "anthropic",
-    defaultModel: "claude-opus-4-6",
+    defaultProvider: "openai",
+    defaultModel: "gpt-5.4",
   });
 }
 
@@ -800,7 +800,7 @@ describe("model-selection", () => {
     it("should fall back to hardcoded default when no custom providers have models", () => {
       const cfg = createProviderWithModelsConfig("empty-provider", []);
       const result = resolveConfiguredRefForTest(cfg);
-      expect(result).toEqual({ provider: "anthropic", model: "claude-opus-4-6" });
+      expect(result).toEqual({ provider: "openai", model: "gpt-5.4" });
     });
 
     it("should warn when specified model cannot be resolved and falls back to default", () => {
@@ -817,13 +817,13 @@ describe("model-selection", () => {
 
         const result = resolveConfiguredModelRef({
           cfg: cfg as OpenClawConfig,
-          defaultProvider: "anthropic",
-          defaultModel: "claude-opus-4-6",
+          defaultProvider: "openai",
+          defaultModel: "gpt-5.4",
         });
 
-        expect(result).toEqual({ provider: "anthropic", model: "claude-opus-4-6" });
+        expect(result).toEqual({ provider: "openai", model: "gpt-5.4" });
         expect(warnSpy).toHaveBeenCalledWith(
-          expect.stringContaining('Falling back to default "anthropic/claude-opus-4-6"'),
+          expect.stringContaining('Falling back to default "openai/gpt-5.4"'),
         );
       } finally {
         warnSpy.mockRestore();
@@ -889,10 +889,10 @@ describe("model-selection", () => {
       expect(resolveAnthropicOpusThinking(cfg)).toBe("adaptive");
     });
 
-    it("defaults Anthropic Claude 4.6 models to adaptive", () => {
+    it("falls back to low when no provider thinking hook is active", () => {
       const cfg = {} as OpenClawConfig;
 
-      expect(resolveAnthropicOpusThinking(cfg)).toBe("adaptive");
+      expect(resolveAnthropicOpusThinking(cfg)).toBe("low");
 
       expect(
         resolveThinkingDefault({
@@ -908,7 +908,7 @@ describe("model-selection", () => {
             },
           ],
         }),
-      ).toBe("adaptive");
+      ).toBe("low");
     });
   });
 });

src/agents/simple-completion-runtime.selection.test.ts

@@ -81,8 +81,8 @@ describe("resolveSimpleCompletionSelectionForAgent", () => {
     const selection = resolveSimpleCompletionSelectionForAgent({ cfg, agentId: "main" });
     expect(selection).toEqual(
       expect.objectContaining({
-        provider: "anthropic",
-        modelId: "claude-opus-4-6",
+        provider: "openai",
+        modelId: "gpt-5.4",
       }),
     );
   });

src/commands/status.summary.test.ts

@@ -14,11 +14,11 @@ vi.mock("./status.summary.runtime.js", () => ({
     classifySessionKey: vi.fn(() => "direct"),
     resolveConfiguredStatusModelRef: vi.fn(() => ({
       provider: "openai",
-      model: "gpt-5.2",
+      model: "gpt-5.4",
     })),
     resolveSessionModelRef: vi.fn(() => ({
       provider: "openai",
-      model: "gpt-5.2",
+      model: "gpt-5.4",
     })),
     resolveContextTokensForModel: vi.fn(() => 200_000),
   },
@@ -26,7 +26,7 @@ vi.mock("./status.summary.runtime.js", () => ({
 
 vi.mock("../agents/defaults.js", () => ({
   DEFAULT_CONTEXT_TOKENS: 200_000,
-  DEFAULT_MODEL: "gpt-5.2",
+  DEFAULT_MODEL: "gpt-5.4",
   DEFAULT_PROVIDER: "openai",
 }));
 

src/infra/provider-usage.auth.normalizes-keys.test.ts

@@ -129,7 +129,6 @@ vi.mock("../agents/auth-profiles.js", () => {
 });
 
 const providerRuntimeMocks = vi.hoisted(() => ({
-  resolveProviderUsageAuthWithPluginMock: vi.fn(async (..._args: unknown[]) => null),
   providerRuntimeMock: {
     augmentModelCatalogWithProviderPlugins: vi.fn((catalog: unknown) => catalog),
     buildProviderAuthDoctorHintWithPlugin: vi.fn(() => undefined),
@@ -153,7 +152,6 @@ const providerRuntimeMocks = vi.hoisted(() => ({
     resolveProviderRuntimePlugin: vi.fn(() => undefined),
     resolveProviderStreamFn: vi.fn(() => undefined),
     resolveProviderSyntheticAuthWithPlugin: vi.fn(() => undefined),
-    resolveProviderUsageSnapshotWithPlugin: vi.fn(async () => undefined),
     resolveProviderXHighThinking: vi.fn(() => undefined),
     runProviderDynamicModel: vi.fn(() => undefined),
     wrapProviderStreamFn: vi.fn(() => undefined),
@@ -167,7 +165,6 @@ vi.mock("../plugins/provider-runtime.js", async () => {
   return {
     ...actual,
     ...providerRuntimeMocks.providerRuntimeMock,
-    resolveProviderUsageAuthWithPlugin: providerRuntimeMocks.resolveProviderUsageAuthWithPluginMock,
   };
 });
 
@@ -178,7 +175,6 @@ vi.mock("../plugins/provider-runtime.ts", async () => {
   return {
     ...actual,
     ...providerRuntimeMocks.providerRuntimeMock,
-    resolveProviderUsageAuthWithPlugin: providerRuntimeMocks.resolveProviderUsageAuthWithPluginMock,
   };
 });
 
@@ -224,8 +220,6 @@ describe("resolveProviderAuths key normalization", () => {
     clearRuntimeConfigSnapshot();
     clearConfigCache();
     clearRuntimeAuthProfileStoreSnapshots();
-    providerRuntimeMocks.resolveProviderUsageAuthWithPluginMock.mockReset();
-    providerRuntimeMocks.resolveProviderUsageAuthWithPluginMock.mockResolvedValue(null);
   });
 
   afterEach(() => {

src/infra/provider-usage.auth.ts

@@ -11,7 +11,6 @@ import { normalizeProviderId } from "../agents/model-selection.js";
 import { loadConfig, type OpenClawConfig } from "../config/config.js";
 import { resolveProviderUsageAuthWithPlugin } from "../plugins/provider-runtime.js";
 import { normalizeSecretInput } from "../utils/normalize-secret-input.js";
-import { resolveLegacyPiAgentAccessToken } from "./provider-usage.shared.js";
 import type { UsageProviderId } from "./provider-usage.types.js";
 
 export type ProviderAuth = {
@@ -29,18 +28,6 @@ type UsageAuthState = {
   agentDir?: string;
 };
 
-function parseGoogleUsageToken(apiKey: string): string {
-  try {
-    const parsed = JSON.parse(apiKey) as { token?: unknown };
-    if (typeof parsed?.token === "string") {
-      return parsed.token;
-    }
-  } catch {
-    // ignore
-  }
-  return apiKey;
-}
-
 function resolveProviderApiKeyFromConfigAndStore(params: {
   state: UsageAuthState;
   providerIds: string[];
@@ -183,46 +170,8 @@ async function resolveProviderUsageAuthFallback(params: {
   state: UsageAuthState;
   provider: UsageProviderId;
 }): Promise<ProviderAuth | null> {
-  switch (params.provider) {
-    case "anthropic":
-    case "github-copilot":
-    case "openai-codex":
-      return await resolveOAuthToken(params);
-    case "google-gemini-cli": {
-      const auth = await resolveOAuthToken(params);
-      return auth ? { ...auth, token: parseGoogleUsageToken(auth.token) } : null;
-    }
-    case "zai": {
-      const apiKey = resolveProviderApiKeyFromConfigAndStore({
-        state: params.state,
-        providerIds: ["zai", "z-ai"],
-        envDirect: [params.state.env.ZAI_API_KEY, params.state.env.Z_AI_API_KEY],
-      });
-      if (apiKey) {
-        return { provider: "zai", token: apiKey };
-      }
-      const legacyToken = resolveLegacyPiAgentAccessToken(params.state.env, ["z-ai", "zai"]);
-      return legacyToken ? { provider: "zai", token: legacyToken } : null;
-    }
-    case "minimax": {
-      const apiKey = resolveProviderApiKeyFromConfigAndStore({
-        state: params.state,
-        providerIds: ["minimax"],
-        envDirect: [params.state.env.MINIMAX_CODE_PLAN_KEY, params.state.env.MINIMAX_API_KEY],
-      });
-      return apiKey ? { provider: "minimax", token: apiKey } : null;
-    }
-    case "xiaomi": {
-      const apiKey = resolveProviderApiKeyFromConfigAndStore({
-        state: params.state,
-        providerIds: ["xiaomi"],
-        envDirect: [params.state.env.XIAOMI_API_KEY],
-      });
-      return apiKey ? { provider: "xiaomi", token: apiKey } : null;
-    }
-    default:
-      return null;
-  }
+  void params;
+  return null;
 }
 
 export async function resolveProviderAuths(params: {

src/infra/provider-usage.load.test.ts

@@ -10,16 +10,9 @@ import {
 
 type ProviderAuth = ProviderUsageAuth<typeof loadProviderUsageSummary>;
 
-const resolveProviderUsageSnapshotWithPlugin = vi.hoisted(() => vi.fn(async () => null));
-
-vi.mock("../plugins/provider-runtime.js", () => ({
-  resolveProviderUsageSnapshotWithPlugin,
-}));
-
 describe("provider-usage.load", () => {
   beforeEach(() => {
-    resolveProviderUsageSnapshotWithPlugin.mockReset();
-    resolveProviderUsageSnapshotWithPlugin.mockResolvedValue(null);
+    vi.restoreAllMocks();
   });
 
   it("loads snapshots for copilot gemini codex and xiaomi", async () => {

src/infra/provider-usage.load.ts

@@ -2,13 +2,6 @@ import { loadConfig, type OpenClawConfig } from "../config/config.js";
 import { resolveProviderUsageSnapshotWithPlugin } from "../plugins/provider-runtime.js";
 import { resolveFetch } from "./fetch.js";
 import { type ProviderAuth, resolveProviderAuths } from "./provider-usage.auth.js";
-import {
-  fetchClaudeUsage,
-  fetchCodexUsage,
-  fetchGeminiUsage,
-  fetchMinimaxUsage,
-  fetchZaiUsage,
-} from "./provider-usage.fetch.js";
 import {
   DEFAULT_TIMEOUT_MS,
   ignoredErrors,
@@ -22,97 +15,19 @@ import type {
   UsageSummary,
 } from "./provider-usage.types.js";
 
-async function fetchCopilotUsageFallback(
-  token: string,
-  timeoutMs: number,
-  fetchFn: typeof fetch,
-): Promise<ProviderUsageSnapshot> {
-  const res = await fetchFn("https://api.github.com/copilot_internal/user", {
-    headers: {
-      Authorization: `token ${token}`,
-      "Editor-Version": "vscode/1.96.2",
-      "User-Agent": "GitHubCopilotChat/0.26.7",
-      "X-Github-Api-Version": "2025-04-01",
-    },
-    signal: AbortSignal.timeout(timeoutMs),
-  });
-  if (!res.ok) {
-    return {
-      provider: "github-copilot",
-      displayName: PROVIDER_LABELS["github-copilot"],
-      windows: [],
-      error: `HTTP ${res.status}`,
-    };
-  }
-  const data = (await res.json()) as {
-    quota_snapshots?: {
-      premium_interactions?: { percent_remaining?: number | null };
-      chat?: { percent_remaining?: number | null };
-    };
-    copilot_plan?: string;
-  };
-  const windows = [];
-  const premiumRemaining = data.quota_snapshots?.premium_interactions?.percent_remaining;
-  if (premiumRemaining !== undefined && premiumRemaining !== null) {
-    windows.push({
-      label: "Premium",
-      usedPercent: Math.max(0, Math.min(100, 100 - premiumRemaining)),
-    });
-  }
-  const chatRemaining = data.quota_snapshots?.chat?.percent_remaining;
-  if (chatRemaining !== undefined && chatRemaining !== null) {
-    windows.push({ label: "Chat", usedPercent: Math.max(0, Math.min(100, 100 - chatRemaining)) });
-  }
-  return {
-    provider: "github-copilot",
-    displayName: PROVIDER_LABELS["github-copilot"],
-    windows,
-    plan: data.copilot_plan,
-  };
-}
-
 async function fetchProviderUsageSnapshotFallback(params: {
   auth: ProviderAuth;
   timeoutMs: number;
   fetchFn: typeof fetch;
 }): Promise<ProviderUsageSnapshot> {
-  switch (params.auth.provider) {
-    case "anthropic":
-      return await fetchClaudeUsage(params.auth.token, params.timeoutMs, params.fetchFn);
-    case "github-copilot":
-      return await fetchCopilotUsageFallback(params.auth.token, params.timeoutMs, params.fetchFn);
-    case "google-gemini-cli":
-      return await fetchGeminiUsage(
-        params.auth.token,
-        params.timeoutMs,
-        params.fetchFn,
-        "google-gemini-cli",
-      );
-    case "openai-codex":
-      return await fetchCodexUsage(
-        params.auth.token,
-        params.auth.accountId,
-        params.timeoutMs,
-        params.fetchFn,
-      );
-    case "zai":
-      return await fetchZaiUsage(params.auth.token, params.timeoutMs, params.fetchFn);
-    case "minimax":
-      return await fetchMinimaxUsage(params.auth.token, params.timeoutMs, params.fetchFn);
-    case "xiaomi":
-      return {
-        provider: "xiaomi",
-        displayName: PROVIDER_LABELS.xiaomi,
-        windows: [],
-      };
-    default:
-      return {
-        provider: params.auth.provider,
-        displayName: PROVIDER_LABELS[params.auth.provider],
-        windows: [],
-        error: "Unsupported provider",
-      };
-  }
+  void params.timeoutMs;
+  void params.fetchFn;
+  return {
+    provider: params.auth.provider,
+    displayName: PROVIDER_LABELS[params.auth.provider] ?? params.auth.provider,
+    windows: [],
+    error: "Unsupported provider",
+  };
 }
 
 type UsageSummaryOptions = {

src/infra/provider-usage.test-support.ts

@@ -25,7 +25,7 @@ export async function loadUsageWithAuth<T extends ProviderUsageLoader>(
     now: usageNow,
     auth,
     fetch: mockFetch as unknown as typeof fetch,
-    // These tests exercise the built-in usage fetchers, not provider plugin hooks.
-    config: { plugins: { enabled: false } } as OpenClawConfig,
+    // Keep config minimal; bundled provider usage hooks own the provider-specific fetchers now.
+    config: {} as OpenClawConfig,
   });
 }