@exactpdf/mcp 0.2.11 → 0.2.13

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/README.md CHANGED
@@ -4,13 +4,15 @@
4
4
 
5
5
  Model Context Protocol (stdio) server for **ExactPDF** — production PDF tools for **Cursor**, **Claude Desktop**, **Codex**, and any MCP client.
6
6
 
7
- Use it when an agent needs to inspect, merge, split, rotate, compress, extract, narrate, translate, or export PDFs without writing fragile PDF plumbing.
7
+ Use it when an agent needs to inspect, merge, split, rotate, compress, extract text, export structured Markdown for RAG, or build images-to-PDF workflows without writing fragile PDF plumbing.
8
8
 
9
9
  **Design principle:** every tool should work on first run, explain credit cost before risky work, return a local output path for files, and expose precise HTTP/API errors when something fails.
10
10
 
11
+ Developer quickstart: [https://exactpdf.com/developers](https://exactpdf.com/developers)
12
+
11
13
  ## Setup
12
14
 
13
- 1. Create an API key at [exactpdf.com](https://exactpdf.com) (Max API keys).
15
+ 1. Create an API key at [max.exactpdf.com/account](https://max.exactpdf.com/account). New accounts get **20 free test credits**. Buy one-time credit packs only after the workflow works; no subscription is required for API/MCP.
14
16
  2. Export:
15
17
 
16
18
  ```bash
@@ -55,10 +57,20 @@ Safe first run:
55
57
  2. exactpdf_account
56
58
  3. exactpdf_pdf_info(path="/absolute/path/document.pdf")
57
59
  4. exactpdf_voice_preview(text="Short sample...", voice_style="professional")
58
- 5. exactpdf_estimate_speech_cost(path="/absolute/path/document.pdf", mode="audiobook")
60
+ 5. exactpdf_estimate_speech_cost(path="/absolute/path/document.pdf", mode="audiobook") # planning only while background audio is hardened
59
61
  ```
60
62
 
61
- Document outputs are saved to `EXACTPDF_API_OUTPUT_DIR` or your OS temp directory. Paid document tools consume credits only on successful output. Async speech jobs are metered by generated minutes.
63
+ Document outputs are saved to `EXACTPDF_API_OUTPUT_DIR` or your OS temp directory. Paid document tools consume credits only on successful output.
64
+
65
+ Background PDF speech, audiobook, translate-and-speak, and presentation narration submissions are currently disabled for paid workloads while ExactPDF finishes dedicated worker hardening and queue-completion smoke. `exactpdf_voice_preview` remains available for short free samples, and `exactpdf_estimate_speech_cost` is planning-only until the API is re-enabled.
66
+
67
+ Credit model:
68
+
69
+ - Free probes: checklist, account, PDF info, voice preview, speech cost estimate, and job polling.
70
+ - New accounts: 20 free test credits for API/MCP workflows.
71
+ - Standard document tools: 1 credit only after a successful output.
72
+ - Background audio jobs: planning-only until dedicated worker smoke passes.
73
+ - One-time packs: available in Max Account when you need more; no subscription required for API/MCP.
62
74
 
63
75
  | Tool | Endpoint | Credits |
64
76
  |------|----------|--------:|
@@ -76,14 +88,14 @@ Document outputs are saved to `EXACTPDF_API_OUTPUT_DIR` or your OS temp director
76
88
  | `exactpdf_pdf_structured_markdown` | POST /api/v1/pdf-structured-markdown | 1 |
77
89
  | `exactpdf_estimate_speech_cost` | local estimate | 0 |
78
90
  | `exactpdf_voice_preview` | POST /api/v1/voice-preview | 0 |
79
- | `exactpdf_pdf_to_speech` | POST /api/v1/pdf-to-speech | 1/min |
80
- | `exactpdf_pdf_to_audiobook` | POST /api/v1/pdf-to-audiobook | 1/min |
81
- | `exactpdf_generate_audiobook` | POST /api/v1/generate-audiobook | 1/min |
82
- | `exactpdf_translate_and_speak` | POST /api/v1/translate-and-speak | 3/min |
83
- | `exactpdf_presentation_narration` | POST /api/v1/presentation-narration | 1/min |
84
- | `exactpdf_job_status` | GET /api/v1/jobs/:id | 0 |
85
- | `exactpdf_get_speech_job` | GET /api/v1/speech-jobs/:id | 0 |
86
- | `exactpdf_download_audio` | GET /api/v1/speech-jobs/:id/download | 0 |
91
+ | `exactpdf_pdf_to_speech` | POST /api/v1/pdf-to-speech | disabled / hardening |
92
+ | `exactpdf_pdf_to_audiobook` | POST /api/v1/pdf-to-audiobook | disabled / hardening |
93
+ | `exactpdf_generate_audiobook` | POST /api/v1/generate-audiobook | disabled / hardening |
94
+ | `exactpdf_translate_and_speak` | POST /api/v1/translate-and-speak | disabled / hardening |
95
+ | `exactpdf_presentation_narration` | POST /api/v1/presentation-narration | disabled / hardening |
96
+ | `exactpdf_job_status` | GET /api/v1/jobs/:id | 0 / legacy polling |
97
+ | `exactpdf_get_speech_job` | GET /api/v1/speech-jobs/:id | 0 / legacy polling |
98
+ | `exactpdf_download_audio` | GET /api/v1/speech-jobs/:id/download | 0 / legacy download |
87
99
 
88
100
  ## Tool behavior
89
101
 
@@ -93,48 +105,27 @@ Document outputs are saved to `EXACTPDF_API_OUTPUT_DIR` or your OS temp director
93
105
  | PDF operations | Absolute local PDF paths | Saved PDF/ZIP path plus remaining credits |
94
106
  | Text / Markdown | Absolute local PDF path | JSON text/Markdown, page count, structured content |
95
107
  | Voice preview | Short text, voice style/language | Saved MP3/WAV preview, no credit charge |
96
- | Speech jobs | PDF path, page range, style, callback URL | `job_id`, status URL, pricing estimate |
97
- | Job polling/download | `job_id` | progress/errors/result URL or saved audio/ZIP |
108
+ | Audio planning | Short text, character count, rough PDF size | Voice preview or estimated minutes/credits |
109
+ | Legacy job polling/download | `job_id` | progress/errors/result URL or saved audio/ZIP for jobs created before the hardening gate |
98
110
 
99
- Async audiobook flow:
111
+ Background audio status:
100
112
 
101
113
  ```text
102
- 1. exactpdf_estimate_speech_cost(path="/abs/book.pdf", mode="audiobook")
114
+ 1. exactpdf_estimate_speech_cost(path="/abs/book.pdf", mode="audiobook") # planning only
103
115
  2. exactpdf_voice_preview(text="Read the first paragraph in an audiobook tone", voice_style="audiobook")
104
- 3. exactpdf_generate_audiobook(path="/abs/book.pdf", voice_style="audiobook")
105
- 4. exactpdf_get_speech_job(job_id="...", download=false)
106
- 5. exactpdf_download_audio(job_id="...") after status is succeeded
107
- ```
108
-
109
- `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`.
110
-
111
- Multilingual speech flow:
112
-
113
- ```text
114
- 1. exactpdf_translate_and_speak(path="/abs/training.pdf", target_language="hi")
115
- 2. exactpdf_job_status(job_id="...", download=false)
116
- 3. exactpdf_job_status(job_id="...", download=true) after status is succeeded
117
- ```
118
-
119
- `segmentation_mode="pages"` is useful for presentation narration because each PDF page becomes a trackable audio section.
120
-
121
- Presentation narration flow:
122
-
123
- ```text
124
- 1. exactpdf_presentation_narration(path="/abs/deck.pdf")
125
- 2. exactpdf_job_status(job_id="...", download=true) after status is succeeded
116
+ 3. Do not call exactpdf_generate_audiobook / exactpdf_pdf_to_speech for paid workloads until ExactPDF announces worker smoke has passed.
126
117
  ```
127
118
 
128
- The default ZIP includes one MP3 per PDF page plus `manifest.json` with page numbers and duration seconds.
119
+ The disabled background audio tools intentionally return an MCP error message before submitting work. This protects users from paying for a path that is still being hardened.
129
120
 
130
121
  ## Reliability rules for agent builders
131
122
 
132
123
  - Always call `exactpdf_account` once before paid tools.
133
124
  - Use absolute file paths. Relative paths are client-dependent and may fail.
134
125
  - Use `exactpdf_pdf_info` before large jobs to confirm the file is readable.
135
- - Use `exactpdf_estimate_speech_cost` before speech/audiobook/translation jobs.
136
- - Use `exactpdf_voice_preview` before full narration when quality matters.
137
- - Async jobs are not downloads. Submit, poll, then download after `succeeded`.
126
+ - Use `exactpdf_estimate_speech_cost` for audio planning only until worker smoke passes.
127
+ - Use `exactpdf_voice_preview` before making any narration-quality decision.
128
+ - Do not promise PDF-to-MP3/audiobook output through MCP until background audio is re-enabled.
138
129
  - If a tool returns `isError`, show the HTTP status and response body to the user. The API returns structured error details.
139
130
  - Keep API keys out of prompts, logs, screenshots, and commits.
140
131
 
@@ -151,6 +142,7 @@ The package must exist on the public npm registry before the MCP Registry can ve
151
142
 
152
143
  ## Docs
153
144
 
145
+ - [https://exactpdf.com/developers](https://exactpdf.com/developers)
154
146
  - [https://exactpdf.com/docs/api](https://exactpdf.com/docs/api)
155
147
 
156
148
  ## License
package/dist/run.d.ts CHANGED
@@ -1,8 +1,11 @@
1
1
  #!/usr/bin/env node
2
2
  /**
3
3
  * ExactPDF MCP — stdio server for Cursor / Claude Desktop / Codex.
4
- * Requires EXACTPDF_API_KEY (sk_live_…) from exactpdf.com (Max dashboard → API keys).
4
+ * Requires EXACTPDF_API_KEY (sk_live_…) from max.exactpdf.com/account.
5
+ * New accounts get free test credits; one-time credit packs are available
6
+ * after the first workflow works.
5
7
  *
8
+ * @see https://exactpdf.com/developers
6
9
  * @see https://exactpdf.com/docs/api
7
10
  */
8
11
  export {};
package/dist/run.js CHANGED
@@ -1,8 +1,11 @@
1
1
  #!/usr/bin/env node
2
2
  /**
3
3
  * ExactPDF MCP — stdio server for Cursor / Claude Desktop / Codex.
4
- * Requires EXACTPDF_API_KEY (sk_live_…) from exactpdf.com (Max dashboard → API keys).
4
+ * Requires EXACTPDF_API_KEY (sk_live_…) from max.exactpdf.com/account.
5
+ * New accounts get free test credits; one-time credit packs are available
6
+ * after the first workflow works.
5
7
  *
8
+ * @see https://exactpdf.com/developers
6
9
  * @see https://exactpdf.com/docs/api
7
10
  */
8
11
  import { mkdir, readFile, stat, writeFile } from 'node:fs/promises';
@@ -20,12 +23,13 @@ function outputDir() {
20
23
  function requireKey() {
21
24
  const k = process.env.EXACTPDF_API_KEY?.trim();
22
25
  if (!k) {
23
- throw new Error('EXACTPDF_API_KEY is not set. Create a key at exactpdf.com (Max → API keys), then export EXACTPDF_API_KEY=sk_live_');
26
+ throw new Error('EXACTPDF_API_KEY is not set. Create a key with free test credits at https://max.exactpdf.com/account, then export EXACTPDF_API_KEY=sk_live_…. Quickstart: https://exactpdf.com/developers');
24
27
  }
25
28
  return k;
26
29
  }
27
- const server = new McpServer({ name: 'exactpdf', version: '0.2.11' }, {
28
- instructions: 'ExactPDF turns local PDFs into production outputs for agents: inspect metadata, merge/split/rotate/compress, create portal-ready merge/images workflows, extract text/structured Markdown, create voice previews, and queue async PDF speech/audiobook/translation/presentation narration jobs. Start with exactpdf_first_run_checklist, then exactpdf_account and exactpdf_pdf_info before paid work. Free tools: first-run checklist, account, pdf-info, voice-preview, speech cost estimate. Standard document tools cost 1 credit on success, including launch portal-ready workflows. Async speech jobs are metered by generated minutes. Use absolute local file paths; binary outputs are saved to EXACTPDF_API_OUTPUT_DIR or the OS temp directory. For async jobs, submit first, then poll exactpdf_get_speech_job, then download with exactpdf_download_audio when succeeded.',
30
+ const BACKGROUND_AUDIO_HARDENING_MESSAGE = 'Background PDF speech, audiobook, translate-and-speak, and presentation narration jobs are currently disabled for paid workloads while ExactPDF finishes dedicated worker hardening and queue-completion smoke. Use exactpdf_voice_preview for short free voice samples and exactpdf_estimate_speech_cost for planning only. Do not promise MP3/audiobook delivery through MCP until the API returns enabled status again.';
31
+ const server = new McpServer({ name: 'exactpdf', version: '0.2.13' }, {
32
+ instructions: 'ExactPDF turns local PDFs into production outputs for agents: inspect metadata, merge/split/rotate/compress, create portal-ready merge/images workflows, extract text, export structured Markdown for RAG, and create short voice previews. Start with exactpdf_first_run_checklist, then exactpdf_account and exactpdf_pdf_info before paid work. New users can create a key with free test credits at https://max.exactpdf.com/account; developer quickstart: https://exactpdf.com/developers. Free tools: first-run checklist, account, pdf-info, voice-preview, and speech cost estimate. Standard document tools cost 1 credit on success, including launch portal-ready workflows. Background PDF speech/audiobook/translation/presentation jobs are in production hardening and must be treated as planning-only until worker smoke passes. Use absolute local file paths; binary outputs are saved to EXACTPDF_API_OUTPUT_DIR or the OS temp directory.',
29
33
  });
30
34
  async function saveBinaryFromResponse(res, prefix, fallbackExt) {
31
35
  const buf = Buffer.from(await res.arrayBuffer());
@@ -68,7 +72,8 @@ server.registerTool('exactpdf_first_run_checklist', {
68
72
  ok: true,
69
73
  setup: {
70
74
  required_env: 'EXACTPDF_API_KEY=sk_live_...',
71
- create_key_url: `${BASE}/max/account`,
75
+ create_key_url: 'https://max.exactpdf.com/account',
76
+ developer_quickstart: `${BASE}/developers`,
72
77
  optional_output_dir: 'EXACTPDF_API_OUTPUT_DIR=/absolute/output/folder',
73
78
  docs: `${BASE}/docs/api`,
74
79
  openapi: `${BASE}/openapi.json`,
@@ -76,20 +81,22 @@ server.registerTool('exactpdf_first_run_checklist', {
76
81
  safe_first_run: [
77
82
  'Call exactpdf_account to verify the key and credit balance.',
78
83
  'Call exactpdf_pdf_info with a small absolute PDF path before extraction or conversion.',
79
- 'Call exactpdf_voice_preview with a short text sample before long narration.',
80
- 'Call exactpdf_estimate_speech_cost before audiobook, speech, translation, or presentation narration.',
84
+ 'Call exactpdf_voice_preview with a short text sample before any narration-quality decision.',
85
+ 'Call exactpdf_estimate_speech_cost only for planning; background audio submission is disabled until worker smoke passes.',
81
86
  ],
82
87
  credit_model: {
83
88
  free: ['exactpdf_first_run_checklist', 'exactpdf_account', 'exactpdf_pdf_info', 'exactpdf_voice_preview', 'exactpdf_estimate_speech_cost', 'job polling'],
89
+ starter_trial: 'New accounts get 20 free test credits for API/MCP workflows.',
84
90
  standard_document_tools: '1 credit only after successful output',
91
+ credit_packs: 'Buy one-time credit packs when the workflow works; subscription is not required for API/MCP.',
85
92
  launch_portal_ready_workflows: '1 credit on successful output: exactpdf_merge_and_compress_pdfs, exactpdf_images_to_pdf_and_compress',
86
- speech_audiobook_presentation: '1 credit per generated minute',
87
- translate_and_speak: '3 credits per generated minute',
93
+ speech_audiobook_presentation: 'Planning only while background audio jobs are disabled for production hardening.',
94
+ translate_and_speak: 'Planning only while background audio jobs are disabled for production hardening.',
88
95
  },
89
96
  reliability_rules: [
90
97
  'Use absolute local file paths; relative paths are client-dependent.',
91
98
  'Never paste API keys into prompts, screenshots, commits, or logs.',
92
- 'Async jobs return a job_id first; poll until succeeded, then download.',
99
+ 'Background audio tools intentionally return a hardening/disabled message until worker smoke passes.',
93
100
  'If a tool returns isError, show the HTTP status and response body to the user.',
94
101
  ],
95
102
  }, null, 2),
@@ -529,7 +536,7 @@ server.registerTool('exactpdf_pdf_structured_markdown', {
529
536
  };
530
537
  });
531
538
  server.registerTool('exactpdf_pdf_to_audiobook', {
532
- description: 'Queue an async PDF-to-audiobook job with chapter-aware narration. Metered by generated minutes. Returns job_id/status_url; poll with exactpdf_get_speech_job and download when succeeded.',
539
+ description: 'Planning-only placeholder while background audio workers are hardened. Do not use for paid workloads; use exactpdf_voice_preview and exactpdf_estimate_speech_cost instead.',
533
540
  inputSchema: z.object({
534
541
  path: z.string().describe('Absolute path to a PDF file'),
535
542
  output_format: z.enum(['mp3', 'wav', 'zip']).optional().describe('Audio export format. Default: mp3.'),
@@ -548,58 +555,14 @@ server.registerTool('exactpdf_pdf_to_audiobook', {
548
555
  webhook_secret: z.string().min(12).max(256).optional().describe('Optional webhook HMAC signing secret.'),
549
556
  }),
550
557
  }, async ({ path, output_format, voice_style, voice_id, language, speed, page_range, normalize_pauses, preserve_chapters, pronunciation_fixes, callback_url, webhook_secret, }) => {
551
- const key = requireKey();
552
- const form = new FormData();
553
- form.append('file', new Blob([await readFile(path)], { type: 'application/pdf' }), basename(path));
554
- if (output_format)
555
- form.append('output_format', output_format);
556
- if (voice_style)
557
- form.append('voice_style', voice_style);
558
- if (voice_id)
559
- form.append('voice_id', voice_id);
560
- if (language)
561
- form.append('language', language);
562
- if (typeof speed === 'number')
563
- form.append('speed', String(speed));
564
- if (page_range)
565
- form.append('page_range', page_range);
566
- if (typeof normalize_pauses === 'boolean')
567
- form.append('normalize_pauses', String(normalize_pauses));
568
- if (typeof preserve_chapters === 'boolean')
569
- form.append('preserve_chapters', String(preserve_chapters));
570
- if (pronunciation_fixes)
571
- form.append('pronunciation_fixes', pronunciation_fixes);
572
- if (callback_url)
573
- form.append('callback_url', callback_url);
574
- if (webhook_secret)
575
- form.append('webhook_secret', webhook_secret);
576
- const res = await fetch(`${BASE}/api/v1/pdf-to-audiobook`, {
577
- method: 'POST',
578
- headers: {
579
- Authorization: `Bearer ${key}`,
580
- Accept: 'application/json',
581
- },
582
- body: form,
583
- });
584
- const raw = await res.text();
585
- if (!res.ok) {
586
- return {
587
- content: [
588
- {
589
- type: 'text',
590
- text: `pdf-to-audiobook failed HTTP ${res.status}\n${raw.slice(0, 8000)}`,
591
- },
592
- ],
593
- isError: true,
594
- };
595
- }
596
558
  return {
597
559
  content: [
598
560
  {
599
561
  type: 'text',
600
- text: `Audiobook job submitted HTTP ${res.status}\n${raw.slice(0, 120_000)}`,
562
+ text: `pdf-to-audiobook disabled: ${BACKGROUND_AUDIO_HARDENING_MESSAGE}`,
601
563
  },
602
564
  ],
565
+ isError: true,
603
566
  };
604
567
  });
605
568
  server.registerTool('exactpdf_job_status', {
@@ -655,7 +618,7 @@ server.registerTool('exactpdf_job_status', {
655
618
  };
656
619
  });
657
620
  server.registerTool('exactpdf_translate_and_speak', {
658
- description: 'Queue async translate-and-speak: translate PDF content, preserve structure, then generate native-accent audio. Costs more than standard speech; start with hi/es/fr. Poll with exactpdf_get_speech_job.',
621
+ description: 'Planning-only placeholder while translate-and-speak workers are hardened. Do not use for paid workloads; use estimates for planning only.',
659
622
  inputSchema: z.object({
660
623
  path: z.string().describe('Absolute path to a PDF file'),
661
624
  target_language: z.string().describe('Target language code, e.g. hi, es, fr, mr, de, ja, ar.'),
@@ -673,59 +636,18 @@ server.registerTool('exactpdf_translate_and_speak', {
673
636
  webhook_secret: z.string().min(12).max(256).optional().describe('Optional webhook HMAC signing secret.'),
674
637
  }),
675
638
  }, async ({ path, target_language, source_language, output_format, voice_style, speed, page_range, segmentation_mode, translation_glossary, callback_url, webhook_secret, }) => {
676
- const key = requireKey();
677
- const form = new FormData();
678
- form.append('file', new Blob([await readFile(path)], { type: 'application/pdf' }), basename(path));
679
- form.append('target_language', target_language);
680
- if (source_language)
681
- form.append('source_language', source_language);
682
- if (output_format)
683
- form.append('output_format', output_format);
684
- if (voice_style)
685
- form.append('voice_style', voice_style);
686
- if (typeof speed === 'number')
687
- form.append('speed', String(speed));
688
- if (page_range)
689
- form.append('page_range', page_range);
690
- if (segmentation_mode)
691
- form.append('segmentation_mode', segmentation_mode);
692
- if (translation_glossary)
693
- form.append('translation_glossary', translation_glossary);
694
- if (callback_url)
695
- form.append('callback_url', callback_url);
696
- if (webhook_secret)
697
- form.append('webhook_secret', webhook_secret);
698
- const res = await fetch(`${BASE}/api/v1/translate-and-speak`, {
699
- method: 'POST',
700
- headers: {
701
- Authorization: `Bearer ${key}`,
702
- Accept: 'application/json',
703
- },
704
- body: form,
705
- });
706
- const raw = await res.text();
707
- if (!res.ok) {
708
- return {
709
- content: [
710
- {
711
- type: 'text',
712
- text: `translate-and-speak failed HTTP ${res.status}\n${raw.slice(0, 8000)}`,
713
- },
714
- ],
715
- isError: true,
716
- };
717
- }
718
639
  return {
719
640
  content: [
720
641
  {
721
642
  type: 'text',
722
- text: `Multilingual speech job submitted HTTP ${res.status}\n${raw.slice(0, 120_000)}`,
643
+ text: `translate-and-speak disabled: ${BACKGROUND_AUDIO_HARDENING_MESSAGE}`,
723
644
  },
724
645
  ],
646
+ isError: true,
725
647
  };
726
648
  });
727
649
  server.registerTool('exactpdf_presentation_narration', {
728
- description: 'Queue audio-first presentation narration for a PDF deck. Each page becomes a narrated slide track in a ZIP with manifest timings. Metered by generated minutes.',
650
+ description: 'Planning-only placeholder while presentation narration workers are hardened. Do not use for paid workloads; use estimates for planning only.',
729
651
  inputSchema: z.object({
730
652
  path: z.string().describe('Absolute path to a PDF presentation file'),
731
653
  output_format: z.enum(['mp3', 'wav', 'zip']).optional().describe('Default: zip for per-slide tracks and manifest.'),
@@ -739,48 +661,14 @@ server.registerTool('exactpdf_presentation_narration', {
739
661
  webhook_secret: z.string().min(12).max(256).optional().describe('Optional webhook HMAC signing secret.'),
740
662
  }),
741
663
  }, async ({ path, output_format, voice_style, speed, page_range, callback_url, webhook_secret }) => {
742
- const key = requireKey();
743
- const form = new FormData();
744
- form.append('file', new Blob([await readFile(path)], { type: 'application/pdf' }), basename(path));
745
- if (output_format)
746
- form.append('output_format', output_format);
747
- if (voice_style)
748
- form.append('voice_style', voice_style);
749
- if (typeof speed === 'number')
750
- form.append('speed', String(speed));
751
- if (page_range)
752
- form.append('page_range', page_range);
753
- if (callback_url)
754
- form.append('callback_url', callback_url);
755
- if (webhook_secret)
756
- form.append('webhook_secret', webhook_secret);
757
- const res = await fetch(`${BASE}/api/v1/presentation-narration`, {
758
- method: 'POST',
759
- headers: {
760
- Authorization: `Bearer ${key}`,
761
- Accept: 'application/json',
762
- },
763
- body: form,
764
- });
765
- const raw = await res.text();
766
- if (!res.ok) {
767
- return {
768
- content: [
769
- {
770
- type: 'text',
771
- text: `presentation-narration failed HTTP ${res.status}\n${raw.slice(0, 8000)}`,
772
- },
773
- ],
774
- isError: true,
775
- };
776
- }
777
664
  return {
778
665
  content: [
779
666
  {
780
667
  type: 'text',
781
- text: `Presentation narration job submitted HTTP ${res.status}\n${raw.slice(0, 120_000)}`,
668
+ text: `presentation-narration disabled: ${BACKGROUND_AUDIO_HARDENING_MESSAGE}`,
782
669
  },
783
670
  ],
671
+ isError: true,
784
672
  };
785
673
  });
786
674
  const speechJobSchema = z.object({
@@ -801,48 +689,9 @@ const speechJobSchema = z.object({
801
689
  webhook_secret: z.string().min(12).max(256).optional().describe('Optional webhook HMAC signing secret.'),
802
690
  });
803
691
  async function submitSpeechJob(endpoint, args, label) {
804
- const key = requireKey();
805
- const form = new FormData();
806
- form.append('file', new Blob([await readFile(args.path)], { type: 'application/pdf' }), basename(args.path));
807
- if (args.output_format)
808
- form.append('output_format', args.output_format);
809
- if (args.voice_style)
810
- form.append('voice_style', args.voice_style);
811
- if (args.voice_id)
812
- form.append('voice_id', args.voice_id);
813
- if (args.language)
814
- form.append('language', args.language);
815
- if (typeof args.speed === 'number')
816
- form.append('speed', String(args.speed));
817
- if (args.page_range)
818
- form.append('page_range', args.page_range);
819
- if (typeof args.normalize_pauses === 'boolean')
820
- form.append('normalize_pauses', String(args.normalize_pauses));
821
- if (typeof args.preserve_chapters === 'boolean')
822
- form.append('preserve_chapters', String(args.preserve_chapters));
823
- if (args.pronunciation_fixes)
824
- form.append('pronunciation_fixes', args.pronunciation_fixes);
825
- if (args.callback_url)
826
- form.append('callback_url', args.callback_url);
827
- if (args.webhook_secret)
828
- form.append('webhook_secret', args.webhook_secret);
829
- const res = await fetch(`${BASE}${endpoint}`, {
830
- method: 'POST',
831
- headers: {
832
- Authorization: `Bearer ${key}`,
833
- Accept: 'application/json',
834
- },
835
- body: form,
836
- });
837
- const raw = await res.text();
838
- if (!res.ok) {
839
- return {
840
- content: [{ type: 'text', text: `${label} failed HTTP ${res.status}\n${raw.slice(0, 8000)}` }],
841
- isError: true,
842
- };
843
- }
844
692
  return {
845
- content: [{ type: 'text', text: `${label} job submitted HTTP ${res.status}\n${raw.slice(0, 120_000)}` }],
693
+ content: [{ type: 'text', text: `${label} disabled: ${BACKGROUND_AUDIO_HARDENING_MESSAGE}` }],
694
+ isError: true,
846
695
  };
847
696
  }
848
697
  async function pollSpeechJob(jobId, download, downloadEndpoint = false) {
@@ -906,7 +755,7 @@ async function pollSpeechJob(jobId, download, downloadEndpoint = false) {
906
755
  };
907
756
  }
908
757
  server.registerTool('exactpdf_estimate_speech_cost', {
909
- description: 'Estimate generated minutes and credits before submitting a paid PDF speech/audiobook/translation job. Local-only rough estimate; free and safe to call repeatedly.',
758
+ description: 'Estimate generated minutes and credits for planning only while background PDF speech/audiobook/translation jobs are hardened. Local-only rough estimate; free and safe to call repeatedly.',
910
759
  inputSchema: z.object({
911
760
  path: z.string().optional().describe('Absolute path to a PDF file. Used for a rough size-based duration estimate.'),
912
761
  characters: z.number().int().positive().optional().describe('Known narration character count, if already extracted.'),
@@ -941,7 +790,7 @@ server.registerTool('exactpdf_estimate_speech_cost', {
941
790
  current_credit_cost: credits,
942
791
  target_language_applied: selectedMode === 'translate' ? (target_language ?? null) : null,
943
792
  ignored_target_language: selectedMode !== 'translate' && target_language ? target_language : null,
944
- note: 'ExactPDF charges async speech jobs by estimated generated minutes: 1 credit/minute for standard speech/audiobook/presentation, 3 credits/minute for translate-and-speak. A target language only changes pricing when mode is translate. Path-based estimates are rough until the API extracts PDF text.',
793
+ note: 'Planning-only while background audio jobs are disabled for production hardening. Intended pricing after re-enable is 1 credit/minute for standard speech/audiobook/presentation and 3 credits/minute for translate-and-speak. A target language only changes pricing when mode is translate. Path-based estimates are rough until the API extracts PDF text.',
945
794
  },
946
795
  }, null, 2),
947
796
  },
@@ -1000,22 +849,22 @@ server.registerTool('exactpdf_voice_preview', {
1000
849
  };
1001
850
  });
1002
851
  server.registerTool('exactpdf_pdf_to_speech', {
1003
- description: 'Queue async PDF-to-speech for reports, study notes, or accessibility audio. Metered by generated minutes. Returns job_id; poll exactpdf_get_speech_job.',
852
+ description: 'Planning-only placeholder while PDF-to-speech workers are hardened. Do not use for paid workloads; use exactpdf_voice_preview and exactpdf_estimate_speech_cost instead.',
1004
853
  inputSchema: speechJobSchema,
1005
854
  }, async (args) => submitSpeechJob('/api/v1/pdf-to-speech', args, 'pdf-to-speech'));
1006
855
  server.registerTool('exactpdf_generate_audiobook', {
1007
- description: 'Queue async audiobook generation with voice style, page range, chapters, pause normalization, and optional webhook. Metered by generated minutes.',
856
+ description: 'Planning-only placeholder while audiobook workers are hardened. Do not use for paid workloads; use exactpdf_voice_preview and exactpdf_estimate_speech_cost instead.',
1008
857
  inputSchema: speechJobSchema,
1009
858
  }, async (args) => submitSpeechJob('/api/v1/generate-audiobook', args, 'generate-audiobook'));
1010
859
  server.registerTool('exactpdf_get_speech_job', {
1011
- description: 'Poll a speech/audiobook/translation/presentation job. Free. Returns status, progress, error details, and result_url when succeeded; set download=true to save result_url locally.',
860
+ description: 'Poll a legacy speech/audiobook/translation/presentation job. Background audio submission is currently disabled for paid workloads.',
1012
861
  inputSchema: z.object({
1013
862
  job_id: z.string().describe('ExactPDF speech job id.'),
1014
863
  download: z.boolean().optional().describe('When true, save result_url to EXACTPDF_API_OUTPUT_DIR if job succeeded.'),
1015
864
  }),
1016
865
  }, async ({ job_id, download }) => pollSpeechJob(job_id, Boolean(download)));
1017
866
  server.registerTool('exactpdf_download_audio', {
1018
- description: 'Download a succeeded speech/audiobook job and save the audio/ZIP locally. Free. Use only after exactpdf_get_speech_job reports succeeded.',
867
+ description: 'Download a legacy succeeded speech/audiobook job and save the audio/ZIP locally. New background audio submission is currently disabled for paid workloads.',
1019
868
  inputSchema: z.object({
1020
869
  job_id: z.string().describe('ExactPDF speech job id.'),
1021
870
  }),
package/package.json CHANGED
@@ -1,7 +1,7 @@
1
1
  {
2
2
  "name": "@exactpdf/mcp",
3
- "version": "0.2.11",
4
- "description": "MCP server for ExactPDF — PDF tools, voice previews, async PDF speech/audiobook jobs, polling, and API credits.",
3
+ "version": "0.2.13",
4
+ "description": "MCP server for ExactPDF — agent PDF workflows, structured Markdown/RAG, merge/split/compress, images-to-PDF, voice previews, and API credits.",
5
5
  "mcpName": "com.exactpdf/mcp",
6
6
  "type": "module",
7
7
  "main": "./dist/run.js",
@@ -19,7 +19,7 @@
19
19
  "publishConfig": {
20
20
  "access": "public"
21
21
  },
22
- "homepage": "https://exactpdf.com/docs/api",
22
+ "homepage": "https://exactpdf.com/developers",
23
23
  "bugs": {
24
24
  "url": "https://exactpdf.com/docs/api",
25
25
  "email": "support@exactpdf.com"
@@ -39,10 +39,19 @@
39
39
  "pdf",
40
40
  "exactpdf",
41
41
  "cursor",
42
+ "cursor-mcp",
42
43
  "claude",
43
44
  "claude-desktop",
45
+ "codex",
44
46
  "headless-pdf",
45
- "pdf-api"
47
+ "pdf-api",
48
+ "pdf-mcp",
49
+ "pdf-compression",
50
+ "pdf-to-markdown",
51
+ "rag",
52
+ "images-to-pdf",
53
+ "merge-pdf",
54
+ "split-pdf"
46
55
  ],
47
56
  "author": "ExactPDF",
48
57
  "license": "MIT",
package/server.json CHANGED
@@ -2,14 +2,14 @@
2
2
  "$schema": "https://static.modelcontextprotocol.io/schemas/2025-12-11/server.schema.json",
3
3
  "name": "com.exactpdf/mcp",
4
4
  "title": "ExactPDF",
5
- "description": "Agent-facing PDF API: merge, split, rotate, compress, images, metadata, text, Markdown, voice previews, async speech/audiobook, multilingual speech, and presentation narration jobs.",
6
- "version": "0.2.11",
5
+ "description": "Agent-facing PDF API: merge, split, rotate, compress, images-to-PDF, metadata, text extraction, structured Markdown/RAG, voice previews, and API credit checks. Background audio jobs are hardening-only until worker smoke passes.",
6
+ "version": "0.2.13",
7
7
  "websiteUrl": "https://exactpdf.com/docs/api",
8
8
  "packages": [
9
9
  {
10
10
  "registryType": "npm",
11
11
  "identifier": "@exactpdf/mcp",
12
- "version": "0.2.11",
12
+ "version": "0.2.13",
13
13
  "transport": {
14
14
  "type": "stdio"
15
15
  }