@stabgan/openrouter-mcp-multimodal 1.8.2 → 2.0.0

package/dist/tool-handlers/image-utils.js

@@ -1,19 +1,15 @@
 import path from 'path';
 import { promises as fs } from 'fs';
-import dns from 'node:dns/promises';
+import { readEnvInt, isBlockedIPv4 as _isBlockedIPv4, assertUrlSafeForFetch as _assertUrlSafeForFetch, fetchHttpResource, } from './fetch-utils.js';
+// Re-export for backward compatibility (tests import from image-utils)
+export const isBlockedIPv4 = _isBlockedIPv4;
+export const assertUrlSafeForFetch = _assertUrlSafeForFetch;
 const DEFAULT_MAX_DIMENSION = 800;
 const DEFAULT_JPEG_QUALITY = 80;
 const DEFAULT_FETCH_TIMEOUT_MS = 30_000;
 const DEFAULT_MAX_DOWNLOAD_BYTES = 25 * 1024 * 1024;
 const DEFAULT_MAX_REDIRECTS = 8;
 const DEFAULT_MAX_DATA_URL_BYTES = 20 * 1024 * 1024;
-function readEnvInt(name, fallback, min = 1) {
-    const raw = process.env[name];
-    if (raw === undefined || raw === '')
-        return fallback;
-    const n = parseInt(raw, 10);
-    return Number.isFinite(n) && n >= min ? n : fallback;
-}
 export function getMaxImageDimension() {
     return readEnvInt('OPENROUTER_IMAGE_MAX_DIMENSION', DEFAULT_MAX_DIMENSION, 64);
 }
@@ -60,152 +56,13 @@ export function getMimeType(filePath) {
     };
     return map[ext] || 'image/jpeg';
 }
-function ipv4ToUint(ip) {
-    const parts = ip.split('.').map((p) => parseInt(p, 10));
-    if (parts.length !== 4 || parts.some((p) => !Number.isInteger(p) || p < 0 || p > 255)) {
-        throw new Error('Invalid IPv4');
-    }
-    return ((parts[0] << 24) | (parts[1] << 16) | (parts[2] << 8) | parts[3]) >>> 0;
-}
-/** Blocks RFC1918, loopback, link-local, CGNAT, metadata (e.g. 169.254.169.254). */
-export function isBlockedIPv4(ip) {
-    const n = ipv4ToUint(ip);
-    if (n >>> 24 === 127)
-        return true;
-    if (n >>> 24 === 10)
-        return true;
-    if (n >>> 20 === 0xac1)
-        return true;
-    if (n >>> 16 === 0xc0a8)
-        return true;
-    if (n >>> 16 === 0xa9fe)
-        return true;
-    if (n >>> 24 === 0)
-        return true;
-    if (n >= 0x64400000 && n <= 0x647fffff)
-        return true;
-    return false;
-}
-function isBlockedIPv6(ip) {
-    const raw = ip.includes('%') ? ip.split('%')[0] : ip;
-    const x = raw.toLowerCase();
-    if (x === '::1')
-        return true;
-    if (x.startsWith('fe80:') || x.startsWith('fec0:'))
-        return true;
-    const first = x.split(':').find((p) => p.length > 0);
-    if (first) {
-        const v = parseInt(first, 16);
-        if (!Number.isNaN(v) && v >= 0xfc00 && v <= 0xfdff)
-            return true;
-    }
-    return false;
-}
-function isIPv4Literal(host) {
-    return /^\d{1,3}(\.\d{1,3}){3}$/.test(host);
-}
-/** Resolve hostname and ensure the resolved address is not private/link-local. */
-export async function assertUrlSafeForFetch(urlString) {
-    let url;
-    try {
-        url = new URL(urlString);
-    }
-    catch {
-        throw new Error('Invalid URL');
-    }
-    if (url.protocol !== 'http:' && url.protocol !== 'https:') {
-        throw new Error('Only HTTP(S) image URLs are allowed');
-    }
-    if (url.username || url.password) {
-        throw new Error('URL with credentials is not allowed');
-    }
-    const host = url.hostname.toLowerCase();
-    if (host === 'localhost' || host.endsWith('.localhost')) {
-        throw new Error('Blocked host');
-    }
-    if (isIPv4Literal(host)) {
-        if (isBlockedIPv4(host))
-            throw new Error('Blocked host');
-        return url;
-    }
-    if (host.includes(':') && !host.startsWith('[')) {
-        if (isBlockedIPv6(host))
-            throw new Error('Blocked host');
-        return url;
-    }
-    let lookupHost = host;
-    if (host.startsWith('[') && host.endsWith(']')) {
-        lookupHost = host.slice(1, -1);
-        if (isBlockedIPv6(lookupHost))
-            throw new Error('Blocked host');
-        return url;
-    }
-    const records = await dns.lookup(lookupHost, { all: true, verbatim: true });
-    if (!records.length)
-        throw new Error('Could not resolve host');
-    for (const r of records) {
-        const { address, family } = r;
-        if (family === 4) {
-            if (isBlockedIPv4(address))
-                throw new Error('Blocked host');
-        }
-        else if (family === 6) {
-            if (isBlockedIPv6(address))
-                throw new Error('Blocked host');
-        }
-    }
-    return url;
-}
-async function readResponseBodyWithLimit(res, maxBytes) {
-    const reader = res.body?.getReader();
-    if (!reader) {
-        const buf = Buffer.from(await res.arrayBuffer());
-        if (buf.length > maxBytes)
-            throw new Error('Response too large');
-        return buf;
-    }
-    const chunks = [];
-    let total = 0;
-    for (;;) {
-        const { done, value } = await reader.read();
-        if (done)
-            break;
-        total += value.byteLength;
-        if (total > maxBytes)
-            throw new Error('Response too large');
-        chunks.push(Buffer.from(value));
-    }
-    return Buffer.concat(chunks);
-}
 export async function fetchHttpImage(urlString) {
-    const maxBytes = getMaxDownloadBytes();
-    const timeoutMs = getFetchTimeoutMs();
-    const maxRedirects = getMaxRedirects();
-    let current = urlString;
-    for (let hop = 0; hop <= maxRedirects; hop++) {
-        const validated = await assertUrlSafeForFetch(current);
-        const target = validated.href;
-        const controller = new AbortController();
-        const t = setTimeout(() => controller.abort(), timeoutMs);
-        let res;
-        try {
-            res = await fetch(target, { redirect: 'manual', signal: controller.signal });
-        }
-        finally {
-            clearTimeout(t);
-        }
-        if (res.status >= 300 && res.status < 400) {
-            const loc = res.headers.get('location');
-            if (!loc)
-                throw new Error('Redirect without Location header');
-            current = new URL(loc, target).href;
-            continue;
-        }
-        if (!res.ok)
-            throw new Error(`HTTP ${res.status}`);
-        return readResponseBodyWithLimit(res, maxBytes);
-    }
-    throw new Error('Too many redirects');
+    const { buffer } = await fetchHttpResource(urlString, {
+        timeoutMs: getFetchTimeoutMs(),
+        maxBytes: getMaxDownloadBytes(),
+        maxRedirects: getMaxRedirects(),
+    });
+    return buffer;
 }
 export async function fetchImage(source) {
     if (source.startsWith('data:')) {

package/dist/tool-handlers.js

@@ -8,6 +8,8 @@ import { handleSearchModels } from './tool-handlers/search-models.js';
 import { handleGetModelInfo } from './tool-handlers/get-model-info.js';
 import { handleValidateModel } from './tool-handlers/validate-model.js';
 import { handleGenerateImage } from './tool-handlers/generate-image.js';
+import { handleAnalyzeAudio } from './tool-handlers/analyze-audio.js';
+import { handleGenerateAudio } from './tool-handlers/generate-audio.js';
 function wrapToolArgs(a) {
     return { params: { arguments: a ?? {} } };
 }
@@ -112,6 +114,36 @@ export class ToolHandlers {
                         required: ['prompt'],
                     },
                 },
+                {
+                    name: 'analyze_audio',
+                    description: 'Analyze or transcribe an audio file using a multimodal model',
+                    inputSchema: {
+                        type: 'object',
+                        properties: {
+                            audio_path: { type: 'string', description: 'File path, URL, or data URL (base64-encoded audio)' },
+                            question: { type: 'string', description: 'Question or instruction about the audio (default: transcribe)' },
+                            model: { type: 'string' },
+                        },
+                        required: ['audio_path'],
+                    },
+                },
+                {
+                    name: 'generate_audio',
+                    description: 'Generate audio from a text prompt. Conversational models (e.g. openai/gpt-audio) respond in spoken audio. ' +
+                        'Music models (e.g. google/lyria-3-clip-preview) need a structured prompt. ' +
+                        'Output format is auto-detected and file extension is corrected automatically.',
+                    inputSchema: {
+                        type: 'object',
+                        properties: {
+                            prompt: { type: 'string', description: 'Text input' },
+                            model: { type: 'string', description: 'Model ID (default: openai/gpt-audio)' },
+                            voice: { type: 'string', description: 'Voice name (default: alloy)' },
+                            format: { type: 'string', description: 'Requested format: pcm16 (default), mp3, flac, opus' },
+                            save_path: { type: 'string', description: 'Path to save audio file. Extension auto-corrected.' },
+                        },
+                        required: ['prompt'],
+                    },
+                },
             ],
         }));
         server.setRequestHandler(CallToolRequestSchema, async (request) => {
@@ -129,6 +161,10 @@ export class ToolHandlers {
                     return handleValidateModel(wrapToolArgs(args), this.modelCache, this.apiClient);
                 case 'generate_image':
                     return handleGenerateImage(wrapToolArgs(args), this.openai);
+                case 'analyze_audio':
+                    return handleAnalyzeAudio(wrapToolArgs(args), this.openai, this.defaultModel);
+                case 'generate_audio':
+                    return handleGenerateAudio(wrapToolArgs(args), this.openai);
                 default:
                     throw new McpError(ErrorCode.MethodNotFound, `Unknown tool: ${name}`);
             }

package/package.json

@@ -1,8 +1,8 @@
 {
   "name": "@stabgan/openrouter-mcp-multimodal",
-  "version": "1.8.2",
+  "version": "2.0.0",
   "mcpName": "io.github.stabgan/openrouter-multimodal",
-  "description": "MCP server for OpenRouter with text chat, image analysis, and image generation",
+  "description": "MCP server for OpenRouter with text chat, image analysis, image generation, audio analysis, and audio generation",
   "type": "module",
   "main": "dist/index.js",
   "bin": {
@@ -28,7 +28,10 @@
     "ai",
     "llm",
     "vision",
-    "image-analysis"
+    "image-analysis",
+    "audio",
+    "transcription",
+    "text-to-speech"
   ],
   "author": "stabgan",
   "repository": {
@@ -45,13 +48,13 @@
   },
   "dependencies": {
     "@modelcontextprotocol/sdk": "^1.27.1",
+    "dotenv": "^16.4.7",
     "openai": "^4.89.1",
     "sharp": "^0.33.5"
   },
   "devDependencies": {
     "@eslint/js": "^9.39.2",
     "@types/node": "^22.13.14",
-    "dotenv": "^16.4.7",
     "eslint": "^9.39.2",
     "eslint-config-prettier": "^10.1.8",
     "prettier": "^3.7.4",