Obsidian Capture

Capture any web content — LinkedIn, GitHub, Medium, X and more — directly into your Obsidian vault.

---

About

Obsidian Capture is a Chrome extension that clips web content and saves it directly to your Obsidian vault as clean, readable Markdown. No copy-paste, no formatting headaches.

Each platform is handled by a dedicated extractor — LinkedIn posts preserve author, role and images; GitHub repos capture the README and metadata; Medium articles come in with full content and tags. Every captured note gets a #Capture tag and the note title is set from the content itself, never a generic placeholder.

Popup

Edit Selection Mode

---

Features

	Feature	Description
🎯	Smart Capture	One click saves the page directly to your vault with proper Markdown formatting
🌐	Site-Specific Extractors	Dedicated parsers for LinkedIn, GitHub, X, Medium, Behance, Dribbble, Facebook, YouTube and Spotify
✏️	Edit Selection Mode	Visually select or deselect page elements before capturing — fine-tune what goes into your note
🏷️	Auto Tag	Every note starts with #Capture for easy filtering in Obsidian
🖼️	Image Capture	Detects and includes images from posts, project galleries and carousels
📝	Readable Markdown	Post text is structured into paragraphs, lists and breathing sections — not a wall of text
📁	Direct Vault Save	Files are written directly to your Obsidian vault folder via the File System Access API — no plugins required
🔒	No Backend	Everything runs locally in your browser. No servers, no accounts, no data sent anywhere

---

Supported Sites

Site	What it captures
LinkedIn	Posts (from feed or direct URL), profiles, job listings, company pages — including images
GitHub	Repo description, README, stars, forks, language, topics — and issues / PRs
X / Twitter	Single tweets and full threads, including images
Medium	Full article content, author, publication, tags, reading time
Behance	Project title, author, tools, image modules, description
Dribbble	Shot title, author, tags, tools, main image
YouTube	Video title, channel, description
Spotify	Title and metadata
Facebook	Posts from feed or profile
Any site	Generic article extractor as fallback — finds and converts the main content

---

Output Format

Every captured note follows this structure:

#Capture

**Author:** Name
**Desc:** Role or headline
**URL:** https://...

---

Content in clean Markdown — paragraphs, lists, images.

The note filename is set from the content title — the first meaningful line of a post, the article headline, or the repo name. Never "LinkedIn Post" or "Untitled".

---

Installation

Obsidian Capture is a local Chrome extension — no Chrome Web Store listing required.

1. Download or clone the repository

git clone https://github.com/uxdreaming/Obsidian-Capture.git

2. Load in Chrome

1. Open Chrome and go to chrome://extensions
2. Enable Developer mode (toggle in the top right)
3. Click Load unpacked
4. Select the obsidian-web-clipper folder

3. Select your Obsidian vault

1. Click the extension icon in the Chrome toolbar
2. Click Select Vault Folder
3. Navigate to your Obsidian vault and confirm

That's it. From now on, every capture saves directly into your vault root.

---

Usage

Quick capture

1. Open any supported page (LinkedIn post, GitHub repo, Medium article, etc.)
2. Click the Obsidian Capture icon in the toolbar
3. The popup shows the detected site — click Capture
4. The note appears in your Obsidian vault instantly

Edit Selection

Use this when you want to pick exactly what goes into the note:

1. Click Edit selection in the popup
2. The page enters selection mode — all content elements are highlighted
3. Click any element to deselect it (it turns grey)
4. Click again to re-add it
5. Press Capture or hit Enter when ready
6. Press Escape or click Cancel to exit without saving

Capture from the feed

On LinkedIn (and Facebook), you don't need to open the individual post. The extension detects which post is most visible in your viewport and captures that one.

---

How It Works

Click Capture  →  Detect site  →  Run extractor  →  Format Markdown  →  Save to vault
                  (URL match)     (site-specific    (paragraphs,        (File System
                                   or generic)       lists, images)      Access API)

1. Detect — The popup matches the current URL against a list of known sites
2. Extract — The matching extractor runs in the page context, scoping all queries to the relevant container
3. Format — Raw text is structured into readable Markdown: paragraphs, lists, images embedded inline
4. Tag — #Capture is prepended to every note
5. Save — The note is written directly to the vault folder with a clean filename derived from the content

---

Project Structure

obsidian-web-clipper/
├── background/
│   └── service-worker.js     # Handles save requests from Edit Selection mode
├── content/
│   ├── content.js            # All extractors + Edit Selection UI logic
│   └── content.css           # Edit Selection overlay styles
├── lib/
│   ├── turndown.min.js       # HTML → Markdown conversion
│   ├── obsidian-formatter.js # Frontmatter & Markdown helpers
│   ├── storage.js            # Vault handle persistence
│   └── templates.js          # Note templates
├── popup/
│   ├── popup.html            # Extension popup UI
│   ├── popup.js              # Site detection, injection, vault save
│   └── popup.css             # Popup styles
├── options/
│   └── options.html          # Settings page
├── icons/                    # Extension icons (16, 48, 128px)
├── screenshots/              # README screenshots
└── manifest.json             # Chrome Extension Manifest V3

---

Completed Features

	Feature
✅	LinkedIn post capture from feed (most visible post in viewport)
✅	LinkedIn profile, job listing and company page extractors
✅	LinkedIn post images (scoped to post container, no leakage from adjacent posts)
✅	GitHub repo extractor (README, stars, forks, language, topics)
✅	GitHub issue and PR extractor
✅	X / Twitter single tweet and thread extractor
✅	Medium full article extractor
✅	Behance project extractor
✅	Dribbble shot extractor
✅	YouTube and Spotify extractors
✅	Facebook post extractor
✅	Generic article fallback for any site
✅	Edit Selection mode — visual element picker
✅	Turndown-powered HTML → Markdown for clean formatting
✅	Post text structured into paragraphs and lists
✅	#Capture tag on every note
✅	Note title derived from content (never a generic placeholder)
✅	No title duplication inside the note
✅	Direct vault write via File System Access API
✅	Permission-aware save from service worker
✅	No backend, no accounts, fully local

---

License

MIT

---

_{Built with 💜 by uxdreaming}