Language: English · 简体中文
Bring web articles into your own local knowledge base.
Open Context Clipper is an open source Chrome extension for capturing the article you are reading. Preview the extracted text, then download Markdown immediately or save the article to a local queue for later. Use the files in your notes or other local knowledge tools. Extraction and queue storage happen on your device. No account or server is required.
| 1. Capture deliberately | 2. Manage locally | 3. Export freely |
|---|---|---|
| Preview the title, source, and extracted text. Download Markdown directly or save to the queue. | Search your local queue, inspect articles, add notes, and select multiple captures. | Download Markdown, bundle articles into a ZIP, or export selected records as JSON. |
Click the extension on an article page to inspect the extracted content and any warnings. Choose Export Markdown to download the current article without adding it to the queue, or Save to queue to keep it for later. An optional note is included in a direct export. Closing the popup saves nothing.
Choose Open queue to search your captures, inspect details, reopen a source, or export on demand. Articles waiting for export stay in Inbox, with the newest captures first. Once a download starts, they move to Completed, where you can find and export them again. Select several articles or export the current search results as a ZIP.
Exported Markdown includes the article title, source URL, capture time, author and note when present, and extracted body. Markdown files use the page title as their filename where possible; unsupported filename characters are replaced, and same-title articles in a ZIP get a number suffix. Open the files in Typora or another Markdown editor, or add them to your local knowledge base for further organization and backup.
You need desktop Google Chrome, Node.js, and npm. From the repository root:
npm ci
npm run buildOpen chrome://extensions, turn on Developer mode, choose Load unpacked, and select the generated dist/ directory. After a source update, rebuild and reload the extension on that page.
- Open a fully loaded article and click Open Context Clipper in the Chrome toolbar.
- Check the title, source, author, publication date, text length, remote image count, and excerpt. Add an optional note.
- Click Export Markdown for an immediate download that does not create a queue record, or Save to queue to keep the article. If the same page and content were already queued, the extension tells you instead of silently creating another copy.
- In Inbox, find newer captures at the top. Export one article as Markdown, selected articles as ZIP or JSON, or all current search results as ZIP.
- A ZIP contains each article in
articles/and the corresponding records incaptures.json. Move the Markdown files into your own folders or knowledge base.
Chrome internal pages such as chrome:// cannot be captured. The popup still lets you open the queue. On sign-in or paywalled pages, the extension works only with content already rendered in your current tab; review the preview before saving.
The interface supports Simplified Chinese and English. It follows Chrome's interface language by default; other browser languages use English. Open Queue → Settings → Language to choose Follow browser, 简体中文, or English. The queue updates immediately, and the popup uses the choice the next time it opens.
Fixed metadata labels in exported Markdown follow the interface language. Article titles, content, and your notes remain unchanged; JSON field names stay stable.
Completed captures are kept locally forever by default. In Queue → Settings, you can choose to keep them for 30 or 90 days. Cleanup runs only the next time you open the queue; Inbox captures are never automatically deleted. Re-exporting starts a new retention period from the latest export. Deleted captures cannot be restored in the extension, so check important downloads before enabling cleanup.
- The extension reads only the already rendered content of the current tab when you click it. It does not poll pages in the background or bypass sign-in and paywalls.
- Captures saved to the queue stay in the extension's local IndexedDB. Direct Markdown exports do not create queue records. Article content is not uploaded to Open Context Forge or a project server.
- Downloads start only when you choose an export. The extension cannot confirm that the file was ultimately saved, so check important exports before enabling automatic cleanup.
- It does not read or save cookies, credentials, or browsing history. Markdown keeps links to remote images instead of downloading the image files.
Local queue data belongs to the installed extension ID; uninstalling or loading from a different path may make older data unavailable. There is currently no JSON import. Export captures you want to keep before changing the installation.
The extension converts rendered article text from the current page into portable Markdown. Content that has not loaded, login prompts, and paywall previews can produce incomplete captures. It does not provide cloud sync, automatic scraping, AI organization, or automatic writes to another knowledge base.
npm test
npm run typecheck
npm run buildAutomated tests cover extraction, queue behavior, and exports. Loading and interacting with the extension still require manual testing in Chrome.
Past verification notes are recorded in verification history.
This project is released under the MIT License. Copyright © Shoukai Huang. Third-party dependencies retain their own licenses.