@changgun/youtube-transcript-mcp 0.1.0

package/dist/fetcher.js

@@ -0,0 +1,41 @@
+import { XMLParser } from "fast-xml-parser";
+export async function fetchTranscript(videoId, lang = "en") {
+    // Step 1: HTML에서 INNERTUBE_API_KEY 추출
+    const pageRes = await fetch(`https://www.youtube.com/watch?v=${videoId}`, {
+        headers: { "Accept-Language": "en-US" },
+    });
+    const html = await pageRes.text();
+    const keyMatch = html.match(/"INNERTUBE_API_KEY":\s*"([a-zA-Z0-9_-]+)"/);
+    if (!keyMatch)
+        throw new Error("INNERTUBE_API_KEY를 찾을 수 없습니다.");
+    // Step 2: player API로 captionTracks 요청
+    const playerRes = await fetch(`https://www.youtube.com/youtubei/v1/player?key=${keyMatch[1]}`, {
+        method: "POST",
+        headers: { "Content-Type": "application/json" },
+        body: JSON.stringify({
+            context: {
+                client: { clientName: "ANDROID", clientVersion: "20.10.38" },
+            },
+            videoId,
+        }),
+    });
+    const playerData = await playerRes.json();
+    const tracks = playerData?.captions?.playerCaptionsTracklistRenderer?.captionTracks ?? [];
+    if (tracks.length === 0)
+        throw new Error("자막이 없습니다.");
+    // 수동 자막 우선, 없으면 ASR
+    const manual = tracks.find((t) => t.languageCode === lang && !t.kind);
+    const asr = tracks.find((t) => t.languageCode === lang && t.kind === "asr");
+    const track = manual ?? asr ?? tracks[0];
+    // Step 3: timedtext XML 받아서 파싱
+    const xmlRes = await fetch(track.baseUrl);
+    const xml = await xmlRes.text();
+    const parser = new XMLParser({ ignoreAttributes: false });
+    const parsed = parser.parse(xml);
+    const paragraphs = parsed?.timedtext?.body?.p ?? [];
+    const text = (Array.isArray(paragraphs) ? paragraphs : [paragraphs])
+        .map((p) => (typeof p === "string" ? p : p["#text"] ?? ""))
+        .filter(Boolean)
+        .join(" ");
+    return text;
+}

package/dist/index.js

@@ -0,0 +1,31 @@
+#!/usr/bin/env node
+import { Server } from "@modelcontextprotocol/sdk/server/index.js";
+import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js";
+import { CallToolRequestSchema, ListToolsRequestSchema } from "@modelcontextprotocol/sdk/types.js";
+import { fetchTranscript } from "./fetcher.js";
+const server = new Server({ name: "youtube-transcript-mcp", version: "0.1.0" }, { capabilities: { tools: {} } });
+server.setRequestHandler(ListToolsRequestSchema, async () => ({
+    tools: [
+        {
+            name: "get_transcript",
+            description: "YouTube 영상의 자막(트랜스크립트)을 텍스트로 반환합니다.",
+            inputSchema: {
+                type: "object",
+                properties: {
+                    videoId: { type: "string", description: "YouTube 영상 ID (예: jNQXAC9IVRw)" },
+                    lang: { type: "string", description: "언어 코드 (기본값: en)" },
+                },
+                required: ["videoId"],
+            },
+        },
+    ],
+}));
+server.setRequestHandler(CallToolRequestSchema, async (request) => {
+    const { videoId, lang = "en" } = request.params.arguments;
+    const transcript = await fetchTranscript(videoId, lang);
+    return {
+        content: [{ type: "text", text: transcript }],
+    };
+});
+const transport = new StdioServerTransport();
+await server.connect(transport);

package/package.json

@@ -0,0 +1,26 @@
+{
+  "name": "@changgun/youtube-transcript-mcp",
+  "version": "0.1.0",
+  "type": "module",
+  "bin": {
+    "youtube-transcript-mcp": "dist/index.js"
+  },
+  "files": [
+    "dist"
+  ],
+  "scripts": {
+    "build": "tsc && chmod +x dist/index.js",
+    "dev": "node --loader ts-node/esm src/index.ts"
+  },
+  "dependencies": {
+    "@modelcontextprotocol/sdk": "^1.9.0",
+    "fast-xml-parser": "^5.5.9",
+    "youtubei.js": "^17.0.1"
+  },
+  "devDependencies": {
+    "@types/node": "^20.0.0",
+    "prettier": "^3.8.1",
+    "tsx": "^4.21.0",
+    "typescript": "^5.6.0"
+  }
+}