refactor(media): move deepgram and groq providers into plugins

src/media-understanding/apply.echo-transcript.test.ts

@@ -165,8 +165,8 @@ describe("applyMediaUnderstanding – echo transcript", () => {
     }));
     vi.doMock("./providers/index.js", async (importOriginal) => {
       const actual = await importOriginal<typeof import("./providers/index.js")>();
-      const { deepgramProvider } = await import("./providers/deepgram/index.js");
-      const { groqProvider } = await import("./providers/groq/index.js");
+      const { deepgramProvider } = await import("../../extensions/deepgram/media-understanding-provider.js");
+      const { groqProvider } = await import("../../extensions/groq/media-understanding-provider.js");
       return {
         ...actual,
         buildMediaUnderstandingRegistry: (

src/media-understanding/apply.test.ts

@@ -248,8 +248,8 @@ describe("applyMediaUnderstanding", () => {
     }));
     vi.doMock("./providers/index.js", async (importOriginal) => {
       const actual = await importOriginal<typeof import("./providers/index.js")>();
-      const { deepgramProvider } = await import("./providers/deepgram/index.js");
-      const { groqProvider } = await import("./providers/groq/index.js");
+      const { deepgramProvider } = await import("../../extensions/deepgram/media-understanding-provider.js");
+      const { groqProvider } = await import("../../extensions/groq/media-understanding-provider.js");
       return {
         ...actual,
         buildMediaUnderstandingRegistry: (

src/media-understanding/providers/deepgram/audio.live.test.ts

@@ -1,6 +1,6 @@
 import { describe, expect, it } from "vitest";
 import { isTruthyEnvValue } from "../../../infra/env.js";
-import { transcribeDeepgramAudio } from "./audio.js";
+import { transcribeDeepgramAudio } from "../../../../extensions/deepgram/media-understanding-provider.js";
 
 const DEEPGRAM_KEY = process.env.DEEPGRAM_API_KEY ?? "";
 const DEEPGRAM_MODEL = process.env.DEEPGRAM_MODEL?.trim() || "nova-3";

src/media-understanding/providers/deepgram/audio.test.ts

@@ -4,7 +4,7 @@ import {
   createRequestCaptureJsonFetch,
   installPinnedHostnameTestHooks,
 } from "../audio.test-helpers.js";
-import { transcribeDeepgramAudio } from "./audio.js";
+import { transcribeDeepgramAudio } from "../../../../extensions/deepgram/media-understanding-provider.js";
 
 installPinnedHostnameTestHooks();
 

src/media-understanding/providers/deepgram/audio.ts

@@ -1,79 +0,0 @@
-import type { AudioTranscriptionRequest, AudioTranscriptionResult } from "../../types.js";
-import {
-  assertOkOrThrowHttpError,
-  normalizeBaseUrl,
-  postTranscriptionRequest,
-  requireTranscriptionText,
-} from "../shared.js";
-
-export const DEFAULT_DEEPGRAM_AUDIO_BASE_URL = "https://api.deepgram.com/v1";
-export const DEFAULT_DEEPGRAM_AUDIO_MODEL = "nova-3";
-
-function resolveModel(model?: string): string {
-  const trimmed = model?.trim();
-  return trimmed || DEFAULT_DEEPGRAM_AUDIO_MODEL;
-}
-
-type DeepgramTranscriptResponse = {
-  results?: {
-    channels?: Array<{
-      alternatives?: Array<{
-        transcript?: string;
-      }>;
-    }>;
-  };
-};
-
-export async function transcribeDeepgramAudio(
-  params: AudioTranscriptionRequest,
-): Promise<AudioTranscriptionResult> {
-  const fetchFn = params.fetchFn ?? fetch;
-  const baseUrl = normalizeBaseUrl(params.baseUrl, DEFAULT_DEEPGRAM_AUDIO_BASE_URL);
-  const allowPrivate = Boolean(params.baseUrl?.trim());
-  const model = resolveModel(params.model);
-
-  const url = new URL(`${baseUrl}/listen`);
-  url.searchParams.set("model", model);
-  if (params.language?.trim()) {
-    url.searchParams.set("language", params.language.trim());
-  }
-  if (params.query) {
-    for (const [key, value] of Object.entries(params.query)) {
-      if (value === undefined) {
-        continue;
-      }
-      url.searchParams.set(key, String(value));
-    }
-  }
-
-  const headers = new Headers(params.headers);
-  if (!headers.has("authorization")) {
-    headers.set("authorization", `Token ${params.apiKey}`);
-  }
-  if (!headers.has("content-type")) {
-    headers.set("content-type", params.mime ?? "application/octet-stream");
-  }
-
-  const body = new Uint8Array(params.buffer);
-  const { response: res, release } = await postTranscriptionRequest({
-    url: url.toString(),
-    headers,
-    body,
-    timeoutMs: params.timeoutMs,
-    fetchFn,
-    allowPrivateNetwork: allowPrivate,
-  });
-
-  try {
-    await assertOkOrThrowHttpError(res, "Audio transcription failed");
-
-    const payload = (await res.json()) as DeepgramTranscriptResponse;
-    const transcript = requireTranscriptionText(
-      payload.results?.channels?.[0]?.alternatives?.[0]?.transcript,
-      "Audio transcription response missing transcript",
-    );
-    return { text: transcript, model };
-  } finally {
-    await release();
-  }
-}

src/media-understanding/providers/deepgram/index.ts

@@ -1,8 +0,0 @@
-import type { MediaUnderstandingProvider } from "../../types.js";
-import { transcribeDeepgramAudio } from "./audio.js";
-
-export const deepgramProvider: MediaUnderstandingProvider = {
-  id: "deepgram",
-  capabilities: ["audio"],
-  transcribeAudio: transcribeDeepgramAudio,
-};

src/media-understanding/providers/groq/index.ts

@@ -1,17 +0,0 @@
-import type { MediaUnderstandingProvider } from "../../types.js";
-import { transcribeOpenAiCompatibleAudio } from "../openai-compatible-audio.js";
-
-const DEFAULT_GROQ_AUDIO_BASE_URL = "https://api.groq.com/openai/v1";
-const DEFAULT_GROQ_AUDIO_MODEL = "whisper-large-v3-turbo";
-
-export const groqProvider: MediaUnderstandingProvider = {
-  id: "groq",
-  capabilities: ["audio"],
-  transcribeAudio: (req) =>
-    transcribeOpenAiCompatibleAudio({
-      ...req,
-      baseUrl: req.baseUrl ?? DEFAULT_GROQ_AUDIO_BASE_URL,
-      defaultBaseUrl: DEFAULT_GROQ_AUDIO_BASE_URL,
-      defaultModel: DEFAULT_GROQ_AUDIO_MODEL,
-    }),
-};

src/media-understanding/providers/index.test.ts

@@ -19,17 +19,6 @@ describe("media-understanding provider registry", () => {
     resetPluginRuntimeStateForTest();
   });
 
-  it("keeps core-owned fallback providers registered by default", () => {
-    const registry = buildMediaUnderstandingRegistry();
-    const groqProvider = getMediaUnderstandingProvider("groq", registry);
-    const deepgramProvider = getMediaUnderstandingProvider("deepgram", registry);
-
-    expect(groqProvider?.id).toBe("groq");
-    expect(groqProvider?.capabilities).toEqual(["audio"]);
-    expect(deepgramProvider?.id).toBe("deepgram");
-    expect(deepgramProvider?.capabilities).toEqual(["audio"]);
-  });
-
   it("merges plugin-registered media providers into the active registry", async () => {
     const pluginRegistry = createEmptyPluginRegistry();
     pluginRegistry.mediaUnderstandingProviders.push({
@@ -75,7 +64,7 @@ describe("media-understanding provider registry", () => {
   it("does not load plugins when config is absent and no runtime registry is active", () => {
     const registry = buildMediaUnderstandingRegistry();
 
-    expect([...registry.keys()]).toEqual(["groq", "deepgram"]);
+    expect([...registry.keys()]).toEqual([]);
     expect(loadOpenClawPluginsMock).not.toHaveBeenCalled();
   });
 });

src/media-understanding/providers/index.ts

@@ -3,10 +3,6 @@ import type { OpenClawConfig } from "../../config/config.js";
 import { loadOpenClawPlugins } from "../../plugins/loader.js";
 import { getActivePluginRegistry } from "../../plugins/runtime.js";
 import type { MediaUnderstandingProvider } from "../types.js";
-import { deepgramProvider } from "./deepgram/index.js";
-import { groqProvider } from "./groq/index.js";
-
-const PROVIDERS: MediaUnderstandingProvider[] = [groqProvider, deepgramProvider];
 
 function mergeProviderIntoRegistry(
   registry: Map<string, MediaUnderstandingProvider>,
@@ -37,9 +33,6 @@ export function buildMediaUnderstandingRegistry(
   cfg?: OpenClawConfig,
 ): Map<string, MediaUnderstandingProvider> {
   const registry = new Map<string, MediaUnderstandingProvider>();
-  for (const provider of PROVIDERS) {
-    mergeProviderIntoRegistry(registry, provider);
-  }
   const active = getActivePluginRegistry();
   const activeEntries = active?.mediaUnderstandingProviders ?? [];
   for (const entry of activeEntries) {