diff --git a/extensions/anthropic/index.ts b/extensions/anthropic/index.ts
index a2491dfbd87..aad11b99a5b 100644
--- a/extensions/anthropic/index.ts
+++ b/extensions/anthropic/index.ts
@@ -23,6 +23,7 @@ import {
 import { buildTokenProfileId, validateAnthropicSetupToken } from "../../src/commands/auth-token.js";
 import { applyAuthProfileConfig } from "../../src/commands/onboard-auth.js";
 import { fetchClaudeUsage } from "../../src/infra/provider-usage.fetch.js";
+import { anthropicProvider } from "../../src/media-understanding/providers/anthropic/index.js";
 import { createProviderApiKeyAuthMethod } from "../../src/plugins/provider-api-key-auth.js";
 import type { ProviderAuthResult } from "../../src/plugins/types.js";
 import { normalizeSecretInput } from "../../src/utils/normalize-secret-input.js";
@@ -394,6 +395,7 @@ const anthropicPlugin = {
           profileId: ctx.profileId,
         }),
     });
+    api.registerMediaUnderstandingProvider(anthropicProvider);
   },
 };
 
diff --git a/extensions/google/index.ts b/extensions/google/index.ts
index 59d417e9349..177de77e49d 100644
--- a/extensions/google/index.ts
+++ b/extensions/google/index.ts
@@ -7,6 +7,7 @@ import {
   GOOGLE_GEMINI_DEFAULT_MODEL,
   applyGoogleGeminiModelDefault,
 } from "../../src/commands/google-gemini-model-default.js";
+import { googleProvider } from "../../src/media-understanding/providers/google/index.js";
 import { emptyPluginConfigSchema } from "../../src/plugins/config-schema.js";
 import { createProviderApiKeyAuthMethod } from "../../src/plugins/provider-api-key-auth.js";
 import type { OpenClawPluginApi } from "../../src/plugins/types.js";
@@ -51,6 +52,7 @@ const googlePlugin = {
       isModernModelRef: ({ modelId }) => isModernGoogleModel(modelId),
     });
     registerGoogleGeminiCliProvider(api);
+    api.registerMediaUnderstandingProvider(googleProvider);
     api.registerWebSearchProvider(
       createPluginBackedWebSearchProvider({
         id: "gemini",
diff --git a/extensions/lobster/src/lobster-tool.test.ts b/extensions/lobster/src/lobster-tool.test.ts
index 0ed5c0eda97..cba95624f07 100644
--- a/extensions/lobster/src/lobster-tool.test.ts
+++ b/extensions/lobster/src/lobster-tool.test.ts
@@ -45,6 +45,7 @@ function fakeApi(overrides: Partial<OpenClawPluginApi> = {}): OpenClawPluginApi
     registerService() {},
     registerProvider() {},
     registerSpeechProvider() {},
+    registerMediaUnderstandingProvider() {},
     registerWebSearchProvider() {},
     registerInteractiveHandler() {},
     registerHook() {},
diff --git a/extensions/minimax/index.ts b/extensions/minimax/index.ts
index 9330e9c4651..8325f6bb078 100644
--- a/extensions/minimax/index.ts
+++ b/extensions/minimax/index.ts
@@ -9,6 +9,10 @@ import {
 import { ensureAuthProfileStore, listProfilesForProvider } from "../../src/agents/auth-profiles.js";
 import { MINIMAX_OAUTH_MARKER } from "../../src/agents/model-auth-markers.js";
 import { fetchMinimaxUsage } from "../../src/infra/provider-usage.fetch.js";
+import {
+  minimaxPortalProvider,
+  minimaxProvider,
+} from "../../src/media-understanding/providers/minimax/index.js";
 import { createProviderApiKeyAuthMethod } from "../../src/plugins/provider-api-key-auth.js";
 import { loginMiniMaxPortalOAuth, type MiniMaxRegion } from "./oauth.js";
 import { applyMinimaxApiConfig, applyMinimaxApiConfigCn } from "./onboard.js";
@@ -270,6 +274,8 @@ const minimaxPlugin = {
       ],
       isModernModelRef: ({ modelId }) => isModernMiniMaxModel(modelId),
     });
+    api.registerMediaUnderstandingProvider(minimaxProvider);
+    api.registerMediaUnderstandingProvider(minimaxPortalProvider);
   },
 };
 
diff --git a/extensions/mistral/index.ts b/extensions/mistral/index.ts
index 10211480a29..7e252281555 100644
--- a/extensions/mistral/index.ts
+++ b/extensions/mistral/index.ts
@@ -1,4 +1,5 @@
 import { emptyPluginConfigSchema, type OpenClawPluginApi } from "openclaw/plugin-sdk/core";
+import { mistralProvider } from "../../src/media-understanding/providers/mistral/index.js";
 import { createProviderApiKeyAuthMethod } from "../../src/plugins/provider-api-key-auth.js";
 import { applyMistralConfig, MISTRAL_DEFAULT_MODEL_REF } from "./onboard.js";
 
@@ -50,6 +51,7 @@ const mistralPlugin = {
         ],
       },
     });
+    api.registerMediaUnderstandingProvider(mistralProvider);
   },
 };
 
diff --git a/extensions/moonshot/index.ts b/extensions/moonshot/index.ts
index 09605ccff85..5cf18d96d8b 100644
--- a/extensions/moonshot/index.ts
+++ b/extensions/moonshot/index.ts
@@ -7,6 +7,7 @@ import {
   getScopedCredentialValue,
   setScopedCredentialValue,
 } from "../../src/agents/tools/web-search-plugin-factory.js";
+import { moonshotProvider } from "../../src/media-understanding/providers/moonshot/index.js";
 import { emptyPluginConfigSchema } from "../../src/plugins/config-schema.js";
 import { createProviderApiKeyAuthMethod } from "../../src/plugins/provider-api-key-auth.js";
 import type { OpenClawPluginApi } from "../../src/plugins/types.js";
@@ -99,6 +100,7 @@ const moonshotPlugin = {
         return createMoonshotThinkingWrapper(ctx.streamFn, thinkingType);
       },
     });
+    api.registerMediaUnderstandingProvider(moonshotProvider);
     api.registerWebSearchProvider(
       createPluginBackedWebSearchProvider({
         id: "kimi",
diff --git a/extensions/openai/index.ts b/extensions/openai/index.ts
index cd528f72211..2fd57473693 100644
--- a/extensions/openai/index.ts
+++ b/extensions/openai/index.ts
@@ -1,4 +1,5 @@
 import { emptyPluginConfigSchema, type OpenClawPluginApi } from "openclaw/plugin-sdk/core";
+import { openaiProvider } from "../../src/media-understanding/providers/openai/index.js";
 import { buildOpenAISpeechProvider } from "../../src/tts/providers/openai.js";
 import { buildOpenAICodexProviderPlugin } from "./openai-codex-provider.js";
 import { buildOpenAIProvider } from "./openai-provider.js";
@@ -12,6 +13,7 @@ const openAIPlugin = {
     api.registerProvider(buildOpenAIProvider());
     api.registerProvider(buildOpenAICodexProviderPlugin());
     api.registerSpeechProvider(buildOpenAISpeechProvider());
+    api.registerMediaUnderstandingProvider(openaiProvider);
   },
 };
 
diff --git a/extensions/test-utils/plugin-api.ts b/extensions/test-utils/plugin-api.ts
index 281e151aeb7..82fe818fdec 100644
--- a/extensions/test-utils/plugin-api.ts
+++ b/extensions/test-utils/plugin-api.ts
@@ -16,6 +16,7 @@ export function createTestPluginApi(api: TestPluginApiInput): OpenClawPluginApi
     registerService() {},
     registerProvider() {},
     registerSpeechProvider() {},
+    registerMediaUnderstandingProvider() {},
     registerWebSearchProvider() {},
     registerInteractiveHandler() {},
     registerCommand() {},
diff --git a/extensions/zai/index.ts b/extensions/zai/index.ts
index aee000ec412..f38058dd9e9 100644
--- a/extensions/zai/index.ts
+++ b/extensions/zai/index.ts
@@ -24,6 +24,7 @@ import { applyAuthProfileConfig } from "../../src/commands/onboard-auth.js";
 import type { SecretInput } from "../../src/config/types.secrets.js";
 import { resolveRequiredHomeDir } from "../../src/infra/home-dir.js";
 import { fetchZaiUsage } from "../../src/infra/provider-usage.fetch.js";
+import { zaiProvider } from "../../src/media-understanding/providers/zai/index.js";
 import { normalizeOptionalSecretInput } from "../../src/utils/normalize-secret-input.js";
 import { detectZaiEndpoint, type ZaiEndpointId } from "./detect.js";
 import { applyZaiConfig, applyZaiProviderConfig, ZAI_DEFAULT_MODEL_REF } from "./onboard.js";
@@ -334,6 +335,7 @@ const zaiPlugin = {
       fetchUsageSnapshot: async (ctx) => await fetchZaiUsage(ctx.token, ctx.timeoutMs, ctx.fetchFn),
       isCacheTtlEligible: () => true,
     });
+    api.registerMediaUnderstandingProvider(zaiProvider);
   },
 };
 
diff --git a/src/auto-reply/reply/route-reply.test.ts b/src/auto-reply/reply/route-reply.test.ts
index 5bf5f5c2cec..4c5dd7be889 100644
--- a/src/auto-reply/reply/route-reply.test.ts
+++ b/src/auto-reply/reply/route-reply.test.ts
@@ -92,6 +92,7 @@ const createRegistry = (channels: PluginRegistry["channels"]): PluginRegistry =>
   })),
   providers: [],
   speechProviders: [],
+  mediaUnderstandingProviders: [],
   webSearchProviders: [],
   gatewayHandlers: {},
   httpRoutes: [],
diff --git a/src/commands/channel-setup/plugin-install.test.ts b/src/commands/channel-setup/plugin-install.test.ts
index 5ad6399fa4a..96ca60e2197 100644
--- a/src/commands/channel-setup/plugin-install.test.ts
+++ b/src/commands/channel-setup/plugin-install.test.ts
@@ -338,6 +338,7 @@ describe("ensureChannelSetupPluginInstalled", () => {
       channelIds: [],
       providerIds: [],
       speechProviderIds: [],
+      mediaUnderstandingProviderIds: [],
       webSearchProviderIds: [],
       gatewayMethods: [],
       cliCommands: [],
diff --git a/src/gateway/server-plugins.test.ts b/src/gateway/server-plugins.test.ts
index 58f5c9da4eb..184cb706762 100644
--- a/src/gateway/server-plugins.test.ts
+++ b/src/gateway/server-plugins.test.ts
@@ -30,6 +30,7 @@ const createRegistry = (diagnostics: PluginDiagnostic[]): PluginRegistry => ({
   commands: [],
   providers: [],
   speechProviders: [],
+  mediaUnderstandingProviders: [],
   webSearchProviders: [],
   gatewayHandlers: {},
   httpRoutes: [],
diff --git a/src/gateway/test-helpers.mocks.ts b/src/gateway/test-helpers.mocks.ts
index e05fcc85320..3617bc896bd 100644
--- a/src/gateway/test-helpers.mocks.ts
+++ b/src/gateway/test-helpers.mocks.ts
@@ -147,6 +147,7 @@ const createStubPluginRegistry = (): PluginRegistry => ({
   channelSetups: [],
   providers: [],
   speechProviders: [],
+  mediaUnderstandingProviders: [],
   webSearchProviders: [],
   gatewayHandlers: {},
   httpRoutes: [],
diff --git a/src/media-understanding/providers/index.test.ts b/src/media-understanding/providers/index.test.ts
index 9294d44acd5..3441b3a9a25 100644
--- a/src/media-understanding/providers/index.test.ts
+++ b/src/media-understanding/providers/index.test.ts
@@ -1,7 +1,13 @@
-import { describe, expect, it } from "vitest";
+import { afterEach, describe, expect, it } from "vitest";
+import { createEmptyPluginRegistry } from "../../plugins/registry.js";
+import { setActivePluginRegistry } from "../../plugins/runtime.js";
 import { buildMediaUnderstandingRegistry, getMediaUnderstandingProvider } from "./index.js";
 
 describe("media-understanding provider registry", () => {
+  afterEach(() => {
+    setActivePluginRegistry(createEmptyPluginRegistry());
+  });
+
   it("registers the Mistral provider", () => {
     const registry = buildMediaUnderstandingRegistry();
     const provider = getMediaUnderstandingProvider("mistral", registry);
@@ -32,4 +38,27 @@ describe("media-understanding provider registry", () => {
     expect(provider?.id).toBe("minimax-portal");
     expect(provider?.capabilities).toEqual(["image"]);
   });
+
+  it("merges plugin-registered media providers into the active registry", async () => {
+    const pluginRegistry = createEmptyPluginRegistry();
+    pluginRegistry.mediaUnderstandingProviders.push({
+      pluginId: "google",
+      pluginName: "Google Plugin",
+      source: "test",
+      provider: {
+        id: "google",
+        capabilities: ["image", "audio", "video"],
+        describeImage: async () => ({ text: "plugin image" }),
+        transcribeAudio: async () => ({ text: "plugin audio" }),
+        describeVideo: async () => ({ text: "plugin video" }),
+      },
+    });
+    setActivePluginRegistry(pluginRegistry);
+
+    const registry = buildMediaUnderstandingRegistry();
+    const provider = getMediaUnderstandingProvider("gemini", registry);
+
+    expect(provider?.id).toBe("google");
+    expect(await provider?.describeVideo?.({} as never)).toEqual({ text: "plugin video" });
+  });
 });
diff --git a/src/media-understanding/providers/index.ts b/src/media-understanding/providers/index.ts
index 0ceaa78fd80..6c2e484dbe5 100644
--- a/src/media-understanding/providers/index.ts
+++ b/src/media-understanding/providers/index.ts
@@ -1,4 +1,5 @@
 import { normalizeProviderId } from "../../agents/model-selection.js";
+import { getActivePluginRegistry } from "../../plugins/runtime.js";
 import type { MediaUnderstandingProvider } from "../types.js";
 import { anthropicProvider } from "./anthropic/index.js";
 import { deepgramProvider } from "./deepgram/index.js";
@@ -23,6 +24,22 @@ const PROVIDERS: MediaUnderstandingProvider[] = [
   deepgramProvider,
 ];
 
+function mergeProviderIntoRegistry(
+  registry: Map<string, MediaUnderstandingProvider>,
+  provider: MediaUnderstandingProvider,
+) {
+  const normalizedKey = normalizeMediaProviderId(provider.id);
+  const existing = registry.get(normalizedKey);
+  const merged = existing
+    ? {
+        ...existing,
+        ...provider,
+        capabilities: provider.capabilities ?? existing.capabilities,
+      }
+    : provider;
+  registry.set(normalizedKey, merged);
+}
+
 export function normalizeMediaProviderId(id: string): string {
   const normalized = normalizeProviderId(id);
   if (normalized === "gemini") {
@@ -36,7 +53,10 @@ export function buildMediaUnderstandingRegistry(
 ): Map<string, MediaUnderstandingProvider> {
   const registry = new Map<string, MediaUnderstandingProvider>();
   for (const provider of PROVIDERS) {
-    registry.set(normalizeMediaProviderId(provider.id), provider);
+    mergeProviderIntoRegistry(registry, provider);
+  }
+  for (const entry of getActivePluginRegistry()?.mediaUnderstandingProviders ?? []) {
+    mergeProviderIntoRegistry(registry, entry.provider);
   }
   if (overrides) {
     for (const [key, provider] of Object.entries(overrides)) {
diff --git a/src/plugin-sdk/core.ts b/src/plugin-sdk/core.ts
index 00621521067..13b075e3352 100644
--- a/src/plugin-sdk/core.ts
+++ b/src/plugin-sdk/core.ts
@@ -1,5 +1,6 @@
 export type {
   AnyAgentTool,
+  MediaUnderstandingProviderPlugin,
   OpenClawPluginConfigSchema,
   ProviderDiscoveryContext,
   ProviderCatalogContext,
diff --git a/src/plugin-sdk/index.ts b/src/plugin-sdk/index.ts
index 1e78ee1c7e2..c5ba9d90541 100644
--- a/src/plugin-sdk/index.ts
+++ b/src/plugin-sdk/index.ts
@@ -108,6 +108,7 @@ export { ACP_ERROR_CODES, AcpRuntimeError } from "../acp/runtime/errors.js";
 export type { AcpRuntimeErrorCode } from "../acp/runtime/errors.js";
 export type {
   AnyAgentTool,
+  MediaUnderstandingProviderPlugin,
   OpenClawPluginConfigSchema,
   OpenClawPluginApi,
   OpenClawPluginService,
diff --git a/src/plugins/contracts/loader.contract.test.ts b/src/plugins/contracts/loader.contract.test.ts
index a42c24712ec..874a94a0b5e 100644
--- a/src/plugins/contracts/loader.contract.test.ts
+++ b/src/plugins/contracts/loader.contract.test.ts
@@ -19,6 +19,7 @@ describe("plugin loader contract", () => {
     loadOpenClawPluginsMock.mockReset();
     loadOpenClawPluginsMock.mockReturnValue({
       providers: [],
+      mediaUnderstandingProviders: [],
       webSearchProviders: [],
     });
   });
diff --git a/src/plugins/contracts/registry.contract.test.ts b/src/plugins/contracts/registry.contract.test.ts
index 48da6c3d9a1..06430449808 100644
--- a/src/plugins/contracts/registry.contract.test.ts
+++ b/src/plugins/contracts/registry.contract.test.ts
@@ -1,5 +1,6 @@
 import { describe, expect, it } from "vitest";
 import {
+  mediaUnderstandingProviderContractRegistry,
   pluginRegistrationContractRegistry,
   providerContractRegistry,
   speechProviderContractRegistry,
@@ -35,6 +36,13 @@ function findSpeechProviderForPlugin(pluginId: string) {
   return entry.provider;
 }
 
+function findMediaUnderstandingProviderIdsForPlugin(pluginId: string) {
+  return mediaUnderstandingProviderContractRegistry
+    .filter((entry) => entry.pluginId === pluginId)
+    .map((entry) => entry.provider.id)
+    .toSorted((left, right) => left.localeCompare(right));
+}
+
 function findRegistrationForPlugin(pluginId: string) {
   const entry = pluginRegistrationContractRegistry.find(
     (candidate) => candidate.pluginId === pluginId,
@@ -61,6 +69,11 @@ describe("plugin contract registry", () => {
     expect(ids).toEqual([...new Set(ids)]);
   });
 
+  it("does not duplicate bundled media provider ids", () => {
+    const ids = mediaUnderstandingProviderContractRegistry.map((entry) => entry.provider.id);
+    expect(ids).toEqual([...new Set(ids)]);
+  });
+
   it("keeps multi-provider plugin ownership explicit", () => {
     expect(findProviderIdsForPlugin("google")).toEqual(["google", "google-gemini-cli"]);
     expect(findProviderIdsForPlugin("minimax")).toEqual(["minimax", "minimax-portal"]);
@@ -82,10 +95,24 @@ describe("plugin contract registry", () => {
     expect(findSpeechProviderIdsForPlugin("openai")).toEqual(["openai"]);
   });
 
+  it("keeps bundled media-understanding ownership explicit", () => {
+    expect(findMediaUnderstandingProviderIdsForPlugin("anthropic")).toEqual(["anthropic"]);
+    expect(findMediaUnderstandingProviderIdsForPlugin("google")).toEqual(["google"]);
+    expect(findMediaUnderstandingProviderIdsForPlugin("minimax")).toEqual([
+      "minimax",
+      "minimax-portal",
+    ]);
+    expect(findMediaUnderstandingProviderIdsForPlugin("mistral")).toEqual(["mistral"]);
+    expect(findMediaUnderstandingProviderIdsForPlugin("moonshot")).toEqual(["moonshot"]);
+    expect(findMediaUnderstandingProviderIdsForPlugin("openai")).toEqual(["openai"]);
+    expect(findMediaUnderstandingProviderIdsForPlugin("zai")).toEqual(["zai"]);
+  });
+
   it("keeps bundled provider and web search tool ownership explicit", () => {
     expect(findRegistrationForPlugin("firecrawl")).toMatchObject({
       providerIds: [],
       speechProviderIds: [],
+      mediaUnderstandingProviderIds: [],
       webSearchProviderIds: ["firecrawl"],
       toolNames: ["firecrawl_search", "firecrawl_scrape"],
     });
@@ -95,14 +122,17 @@ describe("plugin contract registry", () => {
     expect(findRegistrationForPlugin("openai")).toMatchObject({
       providerIds: ["openai", "openai-codex"],
       speechProviderIds: ["openai"],
+      mediaUnderstandingProviderIds: ["openai"],
     });
     expect(findRegistrationForPlugin("elevenlabs")).toMatchObject({
       providerIds: [],
       speechProviderIds: ["elevenlabs"],
+      mediaUnderstandingProviderIds: [],
     });
     expect(findRegistrationForPlugin("microsoft")).toMatchObject({
       providerIds: [],
       speechProviderIds: ["microsoft"],
+      mediaUnderstandingProviderIds: [],
     });
   });
 
diff --git a/src/plugins/contracts/registry.ts b/src/plugins/contracts/registry.ts
index 1dc997d7b2e..14dbb17262c 100644
--- a/src/plugins/contracts/registry.ts
+++ b/src/plugins/contracts/registry.ts
@@ -35,7 +35,12 @@ import xaiPlugin from "../../../extensions/xai/index.js";
 import xiaomiPlugin from "../../../extensions/xiaomi/index.js";
 import zaiPlugin from "../../../extensions/zai/index.js";
 import { createCapturedPluginRegistration } from "../../test-utils/plugin-registration.js";
-import type { ProviderPlugin, SpeechProviderPlugin, WebSearchProviderPlugin } from "../types.js";
+import type {
+  MediaUnderstandingProviderPlugin,
+  ProviderPlugin,
+  SpeechProviderPlugin,
+  WebSearchProviderPlugin,
+} from "../types.js";
 
 type RegistrablePlugin = {
   id: string;
@@ -58,10 +63,16 @@ type SpeechProviderContractEntry = {
   provider: SpeechProviderPlugin;
 };
 
+type MediaUnderstandingProviderContractEntry = {
+  pluginId: string;
+  provider: MediaUnderstandingProviderPlugin;
+};
+
 type PluginRegistrationContractEntry = {
   pluginId: string;
   providerIds: string[];
   speechProviderIds: string[];
+  mediaUnderstandingProviderIds: string[];
   webSearchProviderIds: string[];
   toolNames: string[];
 };
@@ -111,6 +122,16 @@ const bundledWebSearchPlugins: Array<RegistrablePlugin & { credentialValue: unkn
 
 const bundledSpeechPlugins: RegistrablePlugin[] = [elevenLabsPlugin, microsoftPlugin, openAIPlugin];
 
+const bundledMediaUnderstandingPlugins: RegistrablePlugin[] = [
+  anthropicPlugin,
+  googlePlugin,
+  minimaxPlugin,
+  mistralPlugin,
+  moonshotPlugin,
+  openAIPlugin,
+  zaiPlugin,
+];
+
 function captureRegistrations(plugin: RegistrablePlugin) {
   const captured = createCapturedPluginRegistration();
   plugin.register(captured.api);
@@ -146,11 +167,23 @@ export const speechProviderContractRegistry: SpeechProviderContractEntry[] =
     }));
   });
 
+export const mediaUnderstandingProviderContractRegistry: MediaUnderstandingProviderContractEntry[] =
+  bundledMediaUnderstandingPlugins.flatMap((plugin) => {
+    const captured = captureRegistrations(plugin);
+    return captured.mediaUnderstandingProviders.map((provider) => ({
+      pluginId: plugin.id,
+      provider,
+    }));
+  });
+
 const bundledPluginRegistrationList = [
   ...new Map(
-    [...bundledProviderPlugins, ...bundledSpeechPlugins, ...bundledWebSearchPlugins].map(
-      (plugin) => [plugin.id, plugin],
-    ),
+    [
+      ...bundledProviderPlugins,
+      ...bundledSpeechPlugins,
+      ...bundledMediaUnderstandingPlugins,
+      ...bundledWebSearchPlugins,
+    ].map((plugin) => [plugin.id, plugin]),
   ).values(),
 ];
 
@@ -161,6 +194,9 @@ export const pluginRegistrationContractRegistry: PluginRegistrationContractEntry
       pluginId: plugin.id,
       providerIds: captured.providers.map((provider) => provider.id),
       speechProviderIds: captured.speechProviders.map((provider) => provider.id),
+      mediaUnderstandingProviderIds: captured.mediaUnderstandingProviders.map(
+        (provider) => provider.id,
+      ),
       webSearchProviderIds: captured.webSearchProviders.map((provider) => provider.id),
       toolNames: captured.tools.map((tool) => tool.name),
     };
diff --git a/src/plugins/hooks.test-helpers.ts b/src/plugins/hooks.test-helpers.ts
index 7954257e714..ea01163d4b0 100644
--- a/src/plugins/hooks.test-helpers.ts
+++ b/src/plugins/hooks.test-helpers.ts
@@ -17,6 +17,9 @@ export function createMockPluginRegistry(
         hookNames: [],
         channelIds: [],
         providerIds: [],
+        speechProviderIds: [],
+        mediaUnderstandingProviderIds: [],
+        webSearchProviderIds: [],
         gatewayMethods: [],
         cliCommands: [],
         services: [],
@@ -35,13 +38,18 @@ export function createMockPluginRegistry(
       source: "test",
     })),
     tools: [],
+    channels: [],
+    channelSetups: [],
+    providers: [],
+    speechProviders: [],
+    mediaUnderstandingProviders: [],
+    webSearchProviders: [],
     httpRoutes: [],
-    channelRegistrations: [],
     gatewayHandlers: {},
     cliRegistrars: [],
     services: [],
-    providers: [],
     commands: [],
+    diagnostics: [],
   } as unknown as PluginRegistry;
 }
 
diff --git a/src/plugins/loader.ts b/src/plugins/loader.ts
index a2e05fc06b9..873fff6b9bf 100644
--- a/src/plugins/loader.ts
+++ b/src/plugins/loader.ts
@@ -495,6 +495,7 @@ function createPluginRecord(params: {
     channelIds: [],
     providerIds: [],
     speechProviderIds: [],
+    mediaUnderstandingProviderIds: [],
     webSearchProviderIds: [],
     gatewayMethods: [],
     cliCommands: [],
diff --git a/src/plugins/registry.ts b/src/plugins/registry.ts
index 231e6f267aa..bad444289ac 100644
--- a/src/plugins/registry.ts
+++ b/src/plugins/registry.ts
@@ -31,6 +31,7 @@ import type {
   OpenClawPluginHttpRouteHandler,
   OpenClawPluginHttpRouteParams,
   OpenClawPluginHookOptions,
+  MediaUnderstandingProviderPlugin,
   ProviderPlugin,
   OpenClawPluginService,
   OpenClawPluginToolContext,
@@ -119,6 +120,14 @@ export type PluginSpeechProviderRegistration = {
   rootDir?: string;
 };
 
+export type PluginMediaUnderstandingProviderRegistration = {
+  pluginId: string;
+  pluginName?: string;
+  provider: MediaUnderstandingProviderPlugin;
+  source: string;
+  rootDir?: string;
+};
+
 export type PluginHookRegistration = {
   pluginId: string;
   entry: HookEntry;
@@ -164,6 +173,7 @@ export type PluginRecord = {
   channelIds: string[];
   providerIds: string[];
   speechProviderIds: string[];
+  mediaUnderstandingProviderIds: string[];
   webSearchProviderIds: string[];
   gatewayMethods: string[];
   cliCommands: string[];
@@ -185,6 +195,7 @@ export type PluginRegistry = {
   channelSetups: PluginChannelSetupRegistration[];
   providers: PluginProviderRegistration[];
   speechProviders: PluginSpeechProviderRegistration[];
+  mediaUnderstandingProviders: PluginMediaUnderstandingProviderRegistration[];
   webSearchProviders: PluginWebSearchProviderRegistration[];
   gatewayHandlers: GatewayRequestHandlers;
   httpRoutes: PluginHttpRouteRegistration[];
@@ -231,6 +242,7 @@ export function createEmptyPluginRegistry(): PluginRegistry {
     channelSetups: [],
     providers: [],
     speechProviders: [],
+    mediaUnderstandingProviders: [],
     webSearchProviders: [],
     gatewayHandlers: {},
     httpRoutes: [],
@@ -593,6 +605,40 @@ export function createPluginRegistry(registryParams: PluginRegistryParams) {
     });
   };
 
+  const registerMediaUnderstandingProvider = (
+    record: PluginRecord,
+    provider: MediaUnderstandingProviderPlugin,
+  ) => {
+    const id = provider.id.trim();
+    if (!id) {
+      pushDiagnostic({
+        level: "error",
+        pluginId: record.id,
+        source: record.source,
+        message: "media provider registration missing id",
+      });
+      return;
+    }
+    const existing = registry.mediaUnderstandingProviders.find((entry) => entry.provider.id === id);
+    if (existing) {
+      pushDiagnostic({
+        level: "error",
+        pluginId: record.id,
+        source: record.source,
+        message: `media provider already registered: ${id} (${existing.pluginId})`,
+      });
+      return;
+    }
+    record.mediaUnderstandingProviderIds.push(id);
+    registry.mediaUnderstandingProviders.push({
+      pluginId: record.id,
+      pluginName: record.name,
+      provider,
+      source: record.source,
+      rootDir: record.rootDir,
+    });
+  };
+
   const registerWebSearchProvider = (record: PluginRecord, provider: WebSearchProviderPlugin) => {
     const id = provider.id.trim();
     if (!id) {
@@ -836,6 +882,10 @@ export function createPluginRegistry(registryParams: PluginRegistryParams) {
         registrationMode === "full"
           ? (provider) => registerSpeechProvider(record, provider)
           : () => {},
+      registerMediaUnderstandingProvider:
+        registrationMode === "full"
+          ? (provider) => registerMediaUnderstandingProvider(record, provider)
+          : () => {},
       registerWebSearchProvider:
         registrationMode === "full"
           ? (provider) => registerWebSearchProvider(record, provider)
@@ -910,6 +960,7 @@ export function createPluginRegistry(registryParams: PluginRegistryParams) {
     registerChannel,
     registerProvider,
     registerSpeechProvider,
+    registerMediaUnderstandingProvider,
     registerWebSearchProvider,
     registerGatewayMethod,
     registerCli,
diff --git a/src/plugins/types.ts b/src/plugins/types.ts
index 0add5cdcf42..23e761940df 100644
--- a/src/plugins/types.ts
+++ b/src/plugins/types.ts
@@ -25,6 +25,7 @@ import type { GatewayRequestHandler } from "../gateway/server-methods/types.js";
 import type { InternalHookHandler } from "../hooks/internal-hooks.js";
 import type { HookEntry } from "../hooks/types.js";
 import type { ProviderUsageSnapshot } from "../infra/provider-usage.types.js";
+import type { MediaUnderstandingProvider } from "../media-understanding/types.js";
 import type { RuntimeEnv } from "../runtime.js";
 import type { RuntimeWebSearchMetadata } from "../secrets/runtime-web-tools.types.js";
 import type {
@@ -881,6 +882,8 @@ export type PluginSpeechProviderEntry = SpeechProviderPlugin & {
   pluginId: string;
 };
 
+export type MediaUnderstandingProviderPlugin = MediaUnderstandingProvider;
+
 export type OpenClawPluginGatewayMethod = {
   method: string;
   handler: GatewayRequestHandler;
@@ -1240,6 +1243,7 @@ export type OpenClawPluginApi = {
   registerService: (service: OpenClawPluginService) => void;
   registerProvider: (provider: ProviderPlugin) => void;
   registerSpeechProvider: (provider: SpeechProviderPlugin) => void;
+  registerMediaUnderstandingProvider: (provider: MediaUnderstandingProviderPlugin) => void;
   registerWebSearchProvider: (provider: WebSearchProviderPlugin) => void;
   registerInteractiveHandler: (registration: PluginInteractiveHandlerRegistration) => void;
   /**
diff --git a/src/test-utils/channel-plugins.ts b/src/test-utils/channel-plugins.ts
index 588c1ca7db6..1283ac9f506 100644
--- a/src/test-utils/channel-plugins.ts
+++ b/src/test-utils/channel-plugins.ts
@@ -27,6 +27,7 @@ export const createTestRegistry = (channels: TestChannelRegistration[] = []): Pl
   })),
   providers: [],
   speechProviders: [],
+  mediaUnderstandingProviders: [],
   webSearchProviders: [],
   gatewayHandlers: {},
   httpRoutes: [],
diff --git a/src/test-utils/plugin-registration.ts b/src/test-utils/plugin-registration.ts
index 6231dedf17b..de8e5422ccf 100644
--- a/src/test-utils/plugin-registration.ts
+++ b/src/test-utils/plugin-registration.ts
@@ -1,5 +1,6 @@
 import type {
   AnyAgentTool,
+  MediaUnderstandingProviderPlugin,
   OpenClawPluginApi,
   ProviderPlugin,
   SpeechProviderPlugin,
@@ -10,6 +11,7 @@ export type CapturedPluginRegistration = {
   api: OpenClawPluginApi;
   providers: ProviderPlugin[];
   speechProviders: SpeechProviderPlugin[];
+  mediaUnderstandingProviders: MediaUnderstandingProviderPlugin[];
   webSearchProviders: WebSearchProviderPlugin[];
   tools: AnyAgentTool[];
 };
@@ -17,12 +19,14 @@ export type CapturedPluginRegistration = {
 export function createCapturedPluginRegistration(): CapturedPluginRegistration {
   const providers: ProviderPlugin[] = [];
   const speechProviders: SpeechProviderPlugin[] = [];
+  const mediaUnderstandingProviders: MediaUnderstandingProviderPlugin[] = [];
   const webSearchProviders: WebSearchProviderPlugin[] = [];
   const tools: AnyAgentTool[] = [];
 
   return {
     providers,
     speechProviders,
+    mediaUnderstandingProviders,
     webSearchProviders,
     tools,
     api: {
@@ -32,6 +36,9 @@ export function createCapturedPluginRegistration(): CapturedPluginRegistration {
       registerSpeechProvider(provider: SpeechProviderPlugin) {
         speechProviders.push(provider);
       },
+      registerMediaUnderstandingProvider(provider: MediaUnderstandingProviderPlugin) {
+        mediaUnderstandingProviders.push(provider);
+      },
       registerWebSearchProvider(provider: WebSearchProviderPlugin) {
         webSearchProviders.push(provider);
       },