@exactpdf/mcp 0.2.6 → 0.2.7

package/README.md

@@ -56,11 +56,12 @@ For a local checkout of this monorepo:
 | `exactpdf_extract_text` | POST /api/v1/extract-text | 1 |
 | `exactpdf_pdf_structured_markdown` | POST /api/v1/pdf-structured-markdown | 1 |
 | `exactpdf_estimate_speech_cost` | local estimate | 0 |
-| `exactpdf_pdf_to_speech` | POST /api/v1/pdf-to-speech | 10 |
-| `exactpdf_pdf_to_audiobook` | POST /api/v1/pdf-to-audiobook | 10 |
-| `exactpdf_generate_audiobook` | POST /api/v1/generate-audiobook | 10 |
-| `exactpdf_translate_and_speak` | POST /api/v1/translate-and-speak | 20 |
-| `exactpdf_presentation_narration` | POST /api/v1/presentation-narration | 10 |
+| `exactpdf_voice_preview` | POST /api/v1/voice-preview | 0 |
+| `exactpdf_pdf_to_speech` | POST /api/v1/pdf-to-speech | 1/min |
+| `exactpdf_pdf_to_audiobook` | POST /api/v1/pdf-to-audiobook | 1/min |
+| `exactpdf_generate_audiobook` | POST /api/v1/generate-audiobook | 1/min |
+| `exactpdf_translate_and_speak` | POST /api/v1/translate-and-speak | 3/min |
+| `exactpdf_presentation_narration` | POST /api/v1/presentation-narration | 1/min |
 | `exactpdf_job_status` | GET /api/v1/jobs/:id | 0 |
 | `exactpdf_get_speech_job` | GET /api/v1/speech-jobs/:id | 0 |
 | `exactpdf_download_audio` | GET /api/v1/speech-jobs/:id/download | 0 |
@@ -69,9 +70,10 @@ Async audiobook flow:
 
 ```text
 1. exactpdf_estimate_speech_cost(path="/abs/book.pdf", mode="audiobook")
-2. exactpdf_generate_audiobook(path="/abs/book.pdf", voice_style="audiobook")
-3. exactpdf_get_speech_job(job_id="...", download=false)
-4. exactpdf_download_audio(job_id="...") after status is succeeded
+2. exactpdf_voice_preview(text="Read the first paragraph in an audiobook tone", voice_style="audiobook")
+3. exactpdf_generate_audiobook(path="/abs/book.pdf", voice_style="audiobook")
+4. exactpdf_get_speech_job(job_id="...", download=false)
+5. exactpdf_download_audio(job_id="...") after status is succeeded
 ```
 
 `exactpdf_pdf_to_speech`, `exactpdf_pdf_to_audiobook`, and `exactpdf_generate_audiobook` accept optional `callback_url` and `webhook_secret`. ExactPDF signs webhook bodies with `x-exactpdf-signature: sha256=<hmac>` over `timestamp.body`.

package/dist/run.js

@@ -24,8 +24,8 @@ function requireKey() {
     }
     return k;
 }
-const server = new McpServer({ name: 'exactpdf', version: '0.2.6' }, {
-    instructions: 'ExactPDF API tools: exactpdf_account + exactpdf_pdf_info (free); merge, split, rotate, compress, images→PDF, extract-text, pdf-structured-markdown (1 credit each on success); async pdf-to-speech/pdf-to-audiobook/generate-audiobook (10 credits), translate-and-speak (20 credits), presentation narration (10 credits), plus speech job polling/download. Set EXACTPDF_API_KEY.',
+const server = new McpServer({ name: 'exactpdf', version: '0.2.7' }, {
+    instructions: 'ExactPDF API tools: exactpdf_account + exactpdf_pdf_info (free); merge, split, rotate, compress, images→PDF, extract-text, pdf-structured-markdown (1 credit each on success); voice-preview (free), async pdf-to-speech/pdf-to-audiobook/generate-audiobook (10 credits), translate-and-speak (20 credits), presentation narration (10 credits), plus speech job polling/download. Set EXACTPDF_API_KEY.',
 });
 async function saveBinaryFromResponse(res, prefix, fallbackExt) {
     const buf = Buffer.from(await res.arrayBuffer());
@@ -793,7 +793,8 @@ server.registerTool('exactpdf_estimate_speech_cost', {
     }
     const estimatedMinutes = minutes ?? Math.max(1, Math.ceil(estimatedChars / 900));
     const selectedMode = target_language ? 'translate' : (mode ?? 'audiobook');
-    const credits = selectedMode === 'translate' ? 20 : 10;
+    const creditsPerMinute = selectedMode === 'translate' ? 3 : 1;
+    const credits = Math.max(1, estimatedMinutes * creditsPerMinute);
     return {
         content: [
             {
@@ -805,14 +806,66 @@ server.registerTool('exactpdf_estimate_speech_cost', {
                         file_bytes: fileBytes,
                         characters: estimatedChars || null,
                         generated_minutes: estimatedMinutes,
+                        credits_per_minute: creditsPerMinute,
                         current_credit_cost: credits,
-                        note: 'ExactPDF currently charges fixed job credits for async speech jobs: 10 for speech/audiobook/presentation, 20 for translate-and-speak. Minute-based pricing is planned.',
+                        note: 'ExactPDF charges async speech jobs by estimated generated minutes: 1 credit/minute for standard speech, 3 credits/minute for translate-and-speak. Path-based estimates are rough until the API extracts PDF text.',
                     },
                 }, null, 2),
             },
         ],
     };
 });
+server.registerTool('exactpdf_voice_preview', {
+    description: 'Generate a short server-side voice preview before submitting a paid PDF speech job. Saves MP3/WAV locally. Does not consume credits.',
+    inputSchema: z.object({
+        text: z.string().optional().describe('Preview text, capped server-side at 600 characters.'),
+        output_format: z.enum(['mp3', 'wav']).optional().describe('Audio preview format. Default: mp3.'),
+        voice_style: z
+            .enum(['professional', 'audiobook', 'educational', 'presenter', 'conversational'])
+            .optional()
+            .describe('Narration style. Default: professional.'),
+        voice_id: z.string().optional().describe('Optional Google Cloud voice name, e.g. en-US-Standard-H.'),
+        language: z.string().optional().describe('Language hint such as en, hi, es, fr. Default: en.'),
+        speed: z.number().min(0.5).max(2).optional().describe('Speech speed multiplier, 0.5-2.0.'),
+    }),
+}, async ({ text, output_format, voice_style, voice_id, language, speed }) => {
+    const key = requireKey();
+    const res = await fetch(`${BASE}/api/v1/voice-preview`, {
+        method: 'POST',
+        headers: {
+            Authorization: `Bearer ${key}`,
+            Accept: '*/*',
+            'Content-Type': 'application/json',
+        },
+        body: JSON.stringify({
+            ...(text ? { text } : {}),
+            ...(output_format ? { output_format } : {}),
+            ...(voice_style ? { voice_style } : {}),
+            ...(voice_id ? { voice_id } : {}),
+            ...(language ? { language } : {}),
+            ...(typeof speed === 'number' ? { speed } : {}),
+        }),
+    });
+    if (!res.ok) {
+        const raw = await res.text();
+        return {
+            content: [{ type: 'text', text: `voice-preview failed HTTP ${res.status}\n${raw.slice(0, 8000)}` }],
+            isError: true,
+        };
+    }
+    const ext = extensionFromContentType(res.headers.get('content-type'), output_format ?? 'mp3');
+    const outPath = await saveBinaryFromResponse(res, 'exactpdf-voice-preview', ext);
+    return {
+        content: [
+            {
+                type: 'text',
+                text: `Voice preview saved: ${outPath}\n` +
+                    `Credits used: ${res.headers.get('x-credits-used') ?? '0'}\n` +
+                    `Provider: ${res.headers.get('x-exactpdf-provider') ?? 'unknown'}`,
+            },
+        ],
+    };
+});
 server.registerTool('exactpdf_pdf_to_speech', {
     description: 'Create an async PDF→speech job via /api/v1/pdf-to-speech (10 credits). Alias of the production audiobook pipeline.',
     inputSchema: speechJobSchema,

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@exactpdf/mcp",
-  "version": "0.2.6",
-  "description": "MCP server for ExactPDF — PDF tools, async PDF to audiobook jobs, polling, and API credits.",
+  "version": "0.2.7",
+  "description": "MCP server for ExactPDF — PDF tools, voice previews, async PDF speech/audiobook jobs, polling, and API credits.",
   "mcpName": "com.exactpdf/mcp",
   "type": "module",
   "main": "./dist/run.js",

package/server.json

@@ -2,14 +2,14 @@
   "$schema": "https://static.modelcontextprotocol.io/schemas/2025-12-11/server.schema.json",
   "name": "com.exactpdf/mcp",
   "title": "ExactPDF",
-  "description": "Agent-facing PDF API: merge, split, rotate, compress, images, metadata, text, Markdown, async speech/audiobook, multilingual speech, and presentation narration jobs.",
-  "version": "0.2.6",
+  "description": "Agent-facing PDF API: merge, split, rotate, compress, images, metadata, text, Markdown, voice previews, async speech/audiobook, multilingual speech, and presentation narration jobs.",
+  "version": "0.2.7",
   "websiteUrl": "https://exactpdf.com/docs/api",
   "packages": [
     {
       "registryType": "npm",
       "identifier": "@exactpdf/mcp",
-      "version": "0.2.6",
+      "version": "0.2.7",
       "transport": {
         "type": "stdio"
       }