Show HN: Sygal – Convert PDFs, scans and web pages to Markdown, locally
Sygal: Local Conversion of PDFs, Scans, and Web Pages to Markdown
Sygal is a powerful, 100% local document-to-markdown converter designed to turn any file into clean, AI-ready data right on your own hardware.
🚀 Launch Special: Get a lifetime license for a one-time payment of €25 (no subscriptions!).
Whether it is a PDF, a complex spreadsheet, a web page, or even a YouTube link, Sygal transforms them into structured Markdown and JSON. This makes your data perfect for LLM prompts, knowledge bases, or automated pipelines.
🛠️ Technical Specifications & Compatibility
- Trial: 14-day free trial (no credit card required).
- OS Support:
macOS 12+andWindows 10+. - Account: No account creation necessary.
- Speed: Average conversion time of per file.
Supported Input Formats
Sygal is designed to read virtually everything you throw at it:
- Documents:
PDF,DOCX,RTF,EPUB,TXT,XML - Data:
XLSX,XLS,CSV,IPYNB - Presentations:
PPTX - Web/Digital:
HTML,URL,Web,YouTube,RSS - Images/Scans:
PNG,JPG,GIF,WEBP,TIFF,HEIC,AVIF, Screenshots - Email:
MSG,EML - Archives:
ZIP
The Conversion Workflow
🎯 Why Clean Output Matters
A raw file is not the same as clean Markdown when processed by an AI. Sygal normalizes disparate formats into a consistent structure.
1. Token Efficiency
Visual content (scans, PDFs, web pages) contains "noise" like pixels and layout markup that AI models still "pay" for in tokens but cannot use. By converting these to text, the content becomes: This means your AI reads only what matters, reducing costs and increasing speed.
2. Pipeline Consistency
Whether you start with an Excel sheet or a PowerPoint deck, the output is always the same. This eliminates the need for format-specific handling in your workflow.
🌟 Feature Set
The Essentials
- Support for 29+ file formats.
- Dual output: Clean Markdown and structured JSON.
- Side-by-side native preview of original vs. result.
- Direct editing of output before exporting.
Advanced Capabilities
- Custom OCR: Use your own API keys for providers like OpenAI, Claude, Mistral, NVIDIA, or Gemini.
- Web Integration: Direct import via URL.
- Local Library: A persistent, full-text searchable archive of all conversions.
- Token Tracking: Real-time token counts displayed per document.
Volume & Power User Tools
- Batch conversion processing.
- Library filtering and search.
- One-click retry for failed files.
Cmd+Vsupport for instant screenshot conversion.
🛡️ Privacy & Security
Your data remains under your control. Only the specific components you authorize reach the network.
| Stays 100% Local 🏠 | May Go Over Network 🌐 |
|---|---|
| Original files (Word, PDF, etc.) | Images/Scans sent to your chosen OCR provider |
| Converted Markdown & JSON | URLs being imported |
| Local library and history | |
| Encrypted API keys |
OCR Provider Comparison
| Provider | Cost Model | Note |
|---|---|---|
| OpenAI | Per token / Resolution | Default option |
| Mistral | Flat price per page | Best value; cheap async mode |
| Claude | Per token | Premium; priciest at scale |
| NVIDIA | Varies | Offers a free starting option |
| Gemini | Varies | Fully supported |
💻 For Developers & AI Builders
Sygal is more than a GUI; it is a full conversion engine. It is ideal for preparing data for RAG pipelines, vector stores, or autonomous agents.
Installation:
npm install -g sygal-cli
The CLI provides strict JSON output, making it compatible with orchestrators such as Hermes or OpenClaw.
👥 Who is Sygal For?
- AI Developers: Create clean datasets for agents without manual cleanup.
- Sensitive Professionals: Legal, HR, and Healthcare workers who cannot upload private contracts or case files to the cloud.
- Researchers: Build a searchable, local library of meeting notes and documentation.
User Testimonials
"I can finally get AI-ready documents without compromising my clients' data." — Executive Assistant
"I feed Sygal's JSON output straight into my AI pipelines. The structure is consistent every time." — Developer
💰 Pricing & Offer
€29 €25 (One-time purchase)
- License: Covers 2 seats.
- Access: Desktop App + CLI.
- Includes: All features, OCR (BYOK), and major version updates.
- Guarantee: 14-day free trial and 14-day refund policy.
[Start Free Trial] (No card required)