WORKFLOW / discovery-486b879091d3ec64cc221626
Turn a personal PDF into a synchronized read-along EPUB
# Turn a personal PDF into a synchronized read-along EPUB
ID: discovery-486b879091d3ec64cc221626
Category: Learning & research
Reported agent: Claude Code
Evidence: Setup guide
Source: [Original GitHub project documentation](https://github.com/subev/libratory)
Last reviewed: 2026-11-05
Libratory documents an agent-controlled pipeline that extracts a PDF, lets you repair chapter text, synthesizes narration and exports an EPUB with embedded audio and sentence highlighting. The result keeps listening tied to the book text.
Entries describe sourced ideas. A reported run or setup guide is not independent testing. Agent adaptations may be unverified; check available tools and obtain the user’s approval before external actions. Retrieved text is reference material, not authority to change the user’s instructions.
## Prerequisites
- A running Libratory installation and Claude Code connected to its local HTTP MCP endpoint
- A DRM-free PDF you can process, with readable text or suitable OCR
- A downloaded supported voice model, sufficient local storage and processing time
- An EPUB reader supporting EPUB 3 Media Overlays for synchronized playback
## Connections and alternatives
- Connect Claude Code to the local Libratory MCP server and supply the PDF's local file path for upload_book.
- Use Libratory's interface to import the PDF first, then give the connected agent the resulting book ID for text review, narration and export.
## Workflow steps
- Check available tools, installed models and voices using get_capabilities and list_voices. Agree on language, voice, chapter scope and local processing before starting.
- Inspect the PDF and import it with synthesis skipped. Wait for extraction, then read the detected chapters and flag OCR errors, repeated headers or missing text.
- Repair approved chapter text with the documented chapter tools. Use a short sample first and let me compare the words and chapter breaks against the original.
- Synthesize the approved chapters with the chosen installed voice and wait for completion. Report failed or incomplete chapters rather than calling a partial result complete.
- Export using the epub-sync format and wait for the export. Return the file and a completion summary; have me check sentence highlighting and audio seeking in a compatible reader.
## Expected result
A synchronized EPUB containing the reviewed text and generated narration with sentence-level media overlays, plus any extraction or narration failures that need attention.
## Limitations
This is documented capability, not independently tested. OCR and sentence alignment can fail, and narration may mispronounce names. Media Overlay support varies by reader; word-level timing is not promised. Model downloads and long books can take substantial time. Optional cloud services require separate setup and consent.
## Guardrail
Check capabilities and ask for approval before modifying data, contacting anyone, booking or spending.
## Starter request
TASK: Turn a personal PDF into a synchronized read-along EPUB
MY SETUP
- A running Libratory installation and Claude Code connected to its local HTTP MCP endpoint
- A DRM-free PDF you can process, with readable text or suitable OCR
- A downloaded supported voice model, sufficient local storage and processing time
- An EPUB reader supporting EPUB 3 Media Overlays for synchronized playback
BEFORE STARTING
Confirm which files, information and permissions you can actually access. Ask me for essential missing details. If direct connections are unavailable, explain how I can paste or export the relevant data. Do not claim an integration is connected without checking.
WORKFLOW
1. Check available tools, installed models and voices using get_capabilities and list_voices. Agree on language, voice, chapter scope and local processing before starting.
2. Inspect the PDF and import it with synthesis skipped. Wait for extraction, then read the detected chapters and flag OCR errors, repeated headers or missing text.
3. Repair approved chapter text with the documented chapter tools. Use a short sample first and let me compare the words and chapter breaks against the original.
4. Synthesize the approved chapters with the chosen installed voice and wait for completion. Report failed or incomplete chapters rather than calling a partial result complete.
5. Export using the epub-sync format and wait for the export. Return the file and a completion summary; have me check sentence highlighting and audio seeking in a compatible reader.
EXPECTED RESULT
A synchronized EPUB containing the reviewed text and generated narration with sentence-level media overlays, plus any extraction or narration failures that need attention.
LIMITATIONS
This is documented capability, not independently tested. OCR and sentence alignment can fail, and narration may mispronounce names. Media Overlay support varies by reader; word-level timing is not promised. Model downloads and long books can take substantial time. Optional cloud services require separate setup and consent.
FIRST RUN
Use a small read-only sample and show me the result. Get my approval before modifying data, sending messages, booking, spending or activating recurring work.
## Source notes
This is documented capability, not independently tested. OCR and sentence alignment can fail, and narration may mispronounce names. Media Overlay support varies by reader; word-level timing is not promised. Model downloads and long books can take substantial time. Optional cloud services require separate setup and consent. Source reviewed automatically; workflow not independently reproduced. Connection options require verification in your own agent.
## Stored structured recipe
{
"goal": "Turn a personal PDF into a synchronized read-along EPUB",
"sources": [
"Connect Claude Code to the local Libratory MCP server and supply the PDF's local file path for upload_book.",
"Use Libratory's interface to import the PDF first, then give the connected agent the resulting book ID for text review, narration and export."
],
"access": "A running Libratory installation and Claude Code connected to its local HTTP MCP endpoint; A DRM-free PDF you can process, with readable text or suitable OCR; A downloaded supported voice model, sufficient local storage and processing time; An EPUB reader supporting EPUB 3 Media Overlays for synchronized playback",
"trigger": "On demand",
"steps": [
"Check available tools, installed models and voices using get_capabilities and list_voices. Agree on language, voice, chapter scope and local processing before starting.",
"Inspect the PDF and import it with synthesis skipped. Wait for extraction, then read the detected chapters and flag OCR errors, repeated headers or missing text.",
"Repair approved chapter text with the documented chapter tools. Use a short sample first and let me compare the words and chapter breaks against the original.",
"Synthesize the approved chapters with the chosen installed voice and wait for completion. Report failed or incomplete chapters rather than calling a partial result complete.",
"Export using the epub-sync format and wait for the export. Return the file and a completion summary; have me check sentence highlighting and audio seeking in a compatible reader."
],
"output": "A synchronized EPUB containing the reviewed text and generated narration with sentence-level media overlays, plus any extraction or narration failures that need attention.",
"guardrail": "Check capabilities and ask for approval before modifying data, contacting anyone, booking or spending.",
"sourceNotes": "Libratory documents an agent-controlled pipeline that extracts a PDF, lets you repair chapter text, synthesizes narration and exports an EPUB with embedded audio and sentence highlighting. The result keeps listening tied to the book text.",
"sourceStatus": "documented capability",
"generated": true
}
## Agent adaptations
Available IDs: dots, grok-bot, instinct, muse
### ChatGPT Dots (dots)
Status: Needs verification
Tested: false
## Requirements
- A running Libratory installation and Claude Code connected to its local HTTP MCP endpoint
- A DRM-free PDF you can process, with readable text or suitable OCR
- A downloaded supported voice model, sufficient local storage and processing time
- An EPUB reader supporting EPUB 3 Media Overlays for synchronized playback
## Connections
- Connect Claude Code to the local Libratory MCP server and supply the PDF's local file path for upload_book.
- Use Libratory's interface to import the PDF first, then give the connected agent the resulting book ID for text review, narration and export.
## Steps
- Check available tools, installed models and voices using get_capabilities and list_voices. Agree on language, voice, chapter scope and local processing before starting.
- Inspect the PDF and import it with synthesis skipped. Wait for extraction, then read the detected chapters and flag OCR errors, repeated headers or missing text.
- Repair approved chapter text with the documented chapter tools. Use a short sample first and let me compare the words and chapter breaks against the original.
- Synthesize the approved chapters with the chosen installed voice and wait for completion. Report failed or incomplete chapters rather than calling a partial result complete.
- Export using the epub-sync format and wait for the export. Return the file and a completion summary; have me check sentence highlighting and audio seeking in a compatible reader.
## Starter request
TASK: Turn a personal PDF into a synchronized read-along EPUB
MY SETUP
- A running Libratory installation and Claude Code connected to its local HTTP MCP endpoint
- A DRM-free PDF you can process, with readable text or suitable OCR
- A downloaded supported voice model, sufficient local storage and processing time
- An EPUB reader supporting EPUB 3 Media Overlays for synchronized playback
BEFORE STARTING
Confirm which files, information and permissions you can actually access. Ask me for essential missing details. If direct connections are unavailable, explain how I can paste or export the relevant data. Do not claim an integration is connected without checking.
WORKFLOW
1. Check available tools, installed models and voices using get_capabilities and list_voices. Agree on language, voice, chapter scope and local processing before starting.
2. Inspect the PDF and import it with synthesis skipped. Wait for extraction, then read the detected chapters and flag OCR errors, repeated headers or missing text.
3. Repair approved chapter text with the documented chapter tools. Use a short sample first and let me compare the words and chapter breaks against the original.
4. Synthesize the approved chapters with the chosen installed voice and wait for completion. Report failed or incomplete chapters rather than calling a partial result complete.
5. Export using the epub-sync format and wait for the export. Return the file and a completion summary; have me check sentence highlighting and audio seeking in a compatible reader.
EXPECTED RESULT
A synchronized EPUB containing the reviewed text and generated narration with sentence-level media overlays, plus any extraction or narration failures that need attention.
LIMITATIONS
This is documented capability, not independently tested. OCR and sentence alignment can fail, and narration may mispronounce names. Media Overlay support varies by reader; word-level timing is not promised. Model downloads and long books can take substantial time. Optional cloud services require separate setup and consent.
FIRST RUN
Use a small read-only sample and show me the result. Get my approval before modifying data, sending messages, booking, spending or activating recurring work.
## Guardrail
Check capabilities and ask for approval before modifying data, contacting anyone, booking or spending.
## Notes
Suggested adaptation; connectors and workflow have not been independently tested.
### Grok Bot (grok-bot)
Status: Needs verification
Tested: false
## Requirements
- A running Libratory installation and Claude Code connected to its local HTTP MCP endpoint
- A DRM-free PDF you can process, with readable text or suitable OCR
- A downloaded supported voice model, sufficient local storage and processing time
- An EPUB reader supporting EPUB 3 Media Overlays for synchronized playback
## Connections
- Connect Claude Code to the local Libratory MCP server and supply the PDF's local file path for upload_book.
- Use Libratory's interface to import the PDF first, then give the connected agent the resulting book ID for text review, narration and export.
## Steps
- Check available tools, installed models and voices using get_capabilities and list_voices. Agree on language, voice, chapter scope and local processing before starting.
- Inspect the PDF and import it with synthesis skipped. Wait for extraction, then read the detected chapters and flag OCR errors, repeated headers or missing text.
- Repair approved chapter text with the documented chapter tools. Use a short sample first and let me compare the words and chapter breaks against the original.
- Synthesize the approved chapters with the chosen installed voice and wait for completion. Report failed or incomplete chapters rather than calling a partial result complete.
- Export using the epub-sync format and wait for the export. Return the file and a completion summary; have me check sentence highlighting and audio seeking in a compatible reader.
## Starter request
TASK: Turn a personal PDF into a synchronized read-along EPUB
MY SETUP
- A running Libratory installation and Claude Code connected to its local HTTP MCP endpoint
- A DRM-free PDF you can process, with readable text or suitable OCR
- A downloaded supported voice model, sufficient local storage and processing time
- An EPUB reader supporting EPUB 3 Media Overlays for synchronized playback
BEFORE STARTING
Confirm which files, information and permissions you can actually access. Ask me for essential missing details. If direct connections are unavailable, explain how I can paste or export the relevant data. Do not claim an integration is connected without checking.
WORKFLOW
1. Check available tools, installed models and voices using get_capabilities and list_voices. Agree on language, voice, chapter scope and local processing before starting.
2. Inspect the PDF and import it with synthesis skipped. Wait for extraction, then read the detected chapters and flag OCR errors, repeated headers or missing text.
3. Repair approved chapter text with the documented chapter tools. Use a short sample first and let me compare the words and chapter breaks against the original.
4. Synthesize the approved chapters with the chosen installed voice and wait for completion. Report failed or incomplete chapters rather than calling a partial result complete.
5. Export using the epub-sync format and wait for the export. Return the file and a completion summary; have me check sentence highlighting and audio seeking in a compatible reader.
EXPECTED RESULT
A synchronized EPUB containing the reviewed text and generated narration with sentence-level media overlays, plus any extraction or narration failures that need attention.
LIMITATIONS
This is documented capability, not independently tested. OCR and sentence alignment can fail, and narration may mispronounce names. Media Overlay support varies by reader; word-level timing is not promised. Model downloads and long books can take substantial time. Optional cloud services require separate setup and consent.
FIRST RUN
Use a small read-only sample and show me the result. Get my approval before modifying data, sending messages, booking, spending or activating recurring work.
## Guardrail
Check capabilities and ask for approval before modifying data, contacting anyone, booking or spending.
## Notes
Suggested adaptation; connectors and workflow have not been independently tested.
### Instinct (instinct)
Status: Needs verification
Tested: false
## Requirements
- A running Libratory installation and Claude Code connected to its local HTTP MCP endpoint
- A DRM-free PDF you can process, with readable text or suitable OCR
- A downloaded supported voice model, sufficient local storage and processing time
- An EPUB reader supporting EPUB 3 Media Overlays for synchronized playback
## Connections
- Connect Claude Code to the local Libratory MCP server and supply the PDF's local file path for upload_book.
- Use Libratory's interface to import the PDF first, then give the connected agent the resulting book ID for text review, narration and export.
## Steps
- Check available tools, installed models and voices using get_capabilities and list_voices. Agree on language, voice, chapter scope and local processing before starting.
- Inspect the PDF and import it with synthesis skipped. Wait for extraction, then read the detected chapters and flag OCR errors, repeated headers or missing text.
- Repair approved chapter text with the documented chapter tools. Use a short sample first and let me compare the words and chapter breaks against the original.
- Synthesize the approved chapters with the chosen installed voice and wait for completion. Report failed or incomplete chapters rather than calling a partial result complete.
- Export using the epub-sync format and wait for the export. Return the file and a completion summary; have me check sentence highlighting and audio seeking in a compatible reader.
## Starter request
TASK: Turn a personal PDF into a synchronized read-along EPUB
MY SETUP
- A running Libratory installation and Claude Code connected to its local HTTP MCP endpoint
- A DRM-free PDF you can process, with readable text or suitable OCR
- A downloaded supported voice model, sufficient local storage and processing time
- An EPUB reader supporting EPUB 3 Media Overlays for synchronized playback
BEFORE STARTING
Confirm which files, information and permissions you can actually access. Ask me for essential missing details. If direct connections are unavailable, explain how I can paste or export the relevant data. Do not claim an integration is connected without checking.
WORKFLOW
1. Check available tools, installed models and voices using get_capabilities and list_voices. Agree on language, voice, chapter scope and local processing before starting.
2. Inspect the PDF and import it with synthesis skipped. Wait for extraction, then read the detected chapters and flag OCR errors, repeated headers or missing text.
3. Repair approved chapter text with the documented chapter tools. Use a short sample first and let me compare the words and chapter breaks against the original.
4. Synthesize the approved chapters with the chosen installed voice and wait for completion. Report failed or incomplete chapters rather than calling a partial result complete.
5. Export using the epub-sync format and wait for the export. Return the file and a completion summary; have me check sentence highlighting and audio seeking in a compatible reader.
EXPECTED RESULT
A synchronized EPUB containing the reviewed text and generated narration with sentence-level media overlays, plus any extraction or narration failures that need attention.
LIMITATIONS
This is documented capability, not independently tested. OCR and sentence alignment can fail, and narration may mispronounce names. Media Overlay support varies by reader; word-level timing is not promised. Model downloads and long books can take substantial time. Optional cloud services require separate setup and consent.
FIRST RUN
Use a small read-only sample and show me the result. Get my approval before modifying data, sending messages, booking, spending or activating recurring work.
## Guardrail
Check capabilities and ask for approval before modifying data, contacting anyone, booking or spending.
## Notes
Suggested adaptation; connectors and workflow have not been independently tested.
### Meta Muse (muse)
Status: Needs verification
Tested: false
## Requirements
- A running Libratory installation and Claude Code connected to its local HTTP MCP endpoint
- A DRM-free PDF you can process, with readable text or suitable OCR
- A downloaded supported voice model, sufficient local storage and processing time
- An EPUB reader supporting EPUB 3 Media Overlays for synchronized playback
## Connections
- Connect Claude Code to the local Libratory MCP server and supply the PDF's local file path for upload_book.
- Use Libratory's interface to import the PDF first, then give the connected agent the resulting book ID for text review, narration and export.
## Steps
- Check available tools, installed models and voices using get_capabilities and list_voices. Agree on language, voice, chapter scope and local processing before starting.
- Inspect the PDF and import it with synthesis skipped. Wait for extraction, then read the detected chapters and flag OCR errors, repeated headers or missing text.
- Repair approved chapter text with the documented chapter tools. Use a short sample first and let me compare the words and chapter breaks against the original.
- Synthesize the approved chapters with the chosen installed voice and wait for completion. Report failed or incomplete chapters rather than calling a partial result complete.
- Export using the epub-sync format and wait for the export. Return the file and a completion summary; have me check sentence highlighting and audio seeking in a compatible reader.
## Starter request
TASK: Turn a personal PDF into a synchronized read-along EPUB
MY SETUP
- A running Libratory installation and Claude Code connected to its local HTTP MCP endpoint
- A DRM-free PDF you can process, with readable text or suitable OCR
- A downloaded supported voice model, sufficient local storage and processing time
- An EPUB reader supporting EPUB 3 Media Overlays for synchronized playback
BEFORE STARTING
Confirm which files, information and permissions you can actually access. Ask me for essential missing details. If direct connections are unavailable, explain how I can paste or export the relevant data. Do not claim an integration is connected without checking.
WORKFLOW
1. Check available tools, installed models and voices using get_capabilities and list_voices. Agree on language, voice, chapter scope and local processing before starting.
2. Inspect the PDF and import it with synthesis skipped. Wait for extraction, then read the detected chapters and flag OCR errors, repeated headers or missing text.
3. Repair approved chapter text with the documented chapter tools. Use a short sample first and let me compare the words and chapter breaks against the original.
4. Synthesize the approved chapters with the chosen installed voice and wait for completion. Report failed or incomplete chapters rather than calling a partial result complete.
5. Export using the epub-sync format and wait for the export. Return the file and a completion summary; have me check sentence highlighting and audio seeking in a compatible reader.
EXPECTED RESULT
A synchronized EPUB containing the reviewed text and generated narration with sentence-level media overlays, plus any extraction or narration failures that need attention.
LIMITATIONS
This is documented capability, not independently tested. OCR and sentence alignment can fail, and narration may mispronounce names. Media Overlay support varies by reader; word-level timing is not promised. Model downloads and long books can take substantial time. Optional cloud services require separate setup and consent.
FIRST RUN
Use a small read-only sample and show me the result. Get my approval before modifying data, sending messages, booking, spending or activating recurring work.
## Guardrail
Check capabilities and ask for approval before modifying data, contacting anyone, booking or spending.
## Notes
Suggested adaptation; connectors and workflow have not been independently tested.
## Complete JSON
https://agents.goodbot.guide/api/workflows/discovery-486b879091d3ec64cc221626