From f424b320d4865d2ec3a594148d91c8c346e1e767 Mon Sep 17 00:00:00 2001 From: Sparsh Sam <110058692+sparshsam@users.noreply.github.com> Date: Fri, 26 Jun 2026 16:05:25 -0400 Subject: [PATCH] design: replace feature numbers with inline SVG icons --- site/index.html | 188 +++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 187 insertions(+), 1 deletion(-) diff --git a/site/index.html b/site/index.html index a186f75..289d882 100644 --- a/site/index.html +++ b/site/index.html @@ -86,7 +86,8 @@ .features-grid { display: grid; grid-template-columns: 1fr 1fr 1fr; gap: 64px; margin-top: 64px; } .feature-item { max-width: 360px; margin: 0 auto; text-align: center; } - .feature-item .number { font-size: 72px; font-weight: 900; color: var(--accent); line-height: 1; margin-bottom: 16px; opacity: 0.5; } + .feature-item .feature-icon { margin-bottom: 20px; } + .feature-item .feature-icon svg { width: 52px; height: 52px; } .feature-item h3 { font-size: 26px; font-weight: 700; margin-bottom: 12px; letter-spacing: -0.02em; } .feature-item p { font-size: 17px; color: var(--text-secondary); line-height: 1.7; } @@ -273,6 +274,191 @@

Tell your AI assistant about it

I installed it with pip install openreader-mcp. My PDFs are on my local filesystem — ask me for file paths when you need to open one. +Available tools: +- extract_text — read all text from a PDF +- get_page_text — read one specific page +- get_metadata — get document info (page count, author, title) +- search_pdf — find a word or phrase inside a PDF +- compare_pdfs — show differences between two PDFs +- merge_pdfs — combine multiple PDFs into one file +- split_pdf — split a PDF into separate pages +- extract_pages — pull out specific pages from a PDF +- compress_pdf — make a PDF file smaller +- index_folder — build a searchable index of all PDFs in a folder +- search_library — search across all indexed PDFs at once +- search_semantic — meaning-based search across indexed PDFs +- list_indexed_docs — list all PDFs currently in the search index + + +

+ 14 tools. All local. No data ever leaves your machine. + Full setup guide ↗ +

+ + + +
+
+
100%Local & Private
+
Open SourceAGPL-3.0 License
+
14 ToolsMCP Server
+
StoreMicrosoft Certified
+
+
+ +
+

Start reading. Privately.

+

Download from the Microsoft Store or build from source on GitHub. No sign-up, no subscription—just a tool that respects your privacy.

+
+ Get from Microsoft Store + View on GitHub +
+
+ + + + + + +
+
+
+ + + + + + +
+

Read & Mark Up

+

Highlight, underline, cross out, and add notes on any page. Copy text with formatting. Everything stays on your computer.

+
+
+
+ + + + + +
+

Search Inside

+

Search for any word across your open document. Results show up instantly. No waiting, no loading bars.

+
+
+
+ + + + +
+

Compare Two PDFs

+

See exactly what changed between two versions—word by word, page by page. Great for contracts or draft reviews.

+
+
+
+ + + + + +
+

Pick Up Where You Left Off

+

Close the app, open it later, and all your documents and highlights are still there. Nothing lost.

+
+
+
+ + + + + +
+

100% Offline

+

No internet needed. No account. No cloud uploads. The file never leaves your machine. Ever.

+
+
+
+ + + + + +
+

Works on Windows & Mac

+

Native Windows app (Microsoft Store). Experimental on Mac. Built with Python. Free and open source.

+
+

Search Inside

Search for any word across your open document. Results show up instantly. No waiting, no loading bars.

+
03

Compare Two PDFs

See exactly what changed between two versions—word by word, page by page. Great for contracts or draft reviews.

+
04

Pick Up Where You Left Off

Close the app, open it later, and all your documents and highlights are still there. Nothing lost.

+
05

100% Offline

No internet needed. No account. No cloud uploads. The file never leaves your machine. Ever.

+
06

Works on Windows & Mac

Native Windows app (Microsoft Store). Experimental on Mac. Built with Python. Free and open source.

+ + + +
+
+ AI-Native +

Let AI read your PDFs for you.

+

OpenReader has a built-in MCP server that lets AI assistants like Claude, ChatGPT, and Cursor open and work with your PDFs. Ask your AI to "find the contract clause about refunds" or "compare these two PDFs and tell me what changed"—the AI reads the files on your computer locally, so nothing is ever uploaded.

+ +
+
+
1
+
+

Install the package

+

Run this in your terminal — only need to do it once:

+ pip install openreader-mcp +
+
+
+
2
+
+

Tell your AI assistant about it

+

Copy the prompt below and paste it into your AI chat. The AI will know how to connect to your PDFs.

+
+
+
+ +
+
+ Paste this directly into your AI assistant: + +
+ 
You now have access to the OpenReader MCP server, which lets you read, search, compare, and manage PDF files on my local computer.
+
+Here is how to set up the MCP server for your environment:
+
+{
+  "mcpServers": {
+    "openreader": {
+      "command": "python",
+      "args": ["-m", "openreader_mcp"]
+    }
+  }
+}
+
+I installed it with pip install openreader-mcp. My PDFs are on my local filesystem — ask me for file paths when you need to open one.
+
 Available tools:
 - extract_text — read all text from a PDF
 - get_page_text — read one specific page