A browser extension that filters unwanted content related to conflict, violence, and inflammatory rhetoric across every website.
⚠️ Official Preview Release: This is a preview version (v0.1.0-preview) of Calm the Chaos. Features may be incomplete or subject to change.By default, content filtering is only enabled for specific platforms (CNN, LinkedIn, Reddit, StackOverflow and YouTube). You can enable filtering for additional domains through the extension settings.
The following categories are defined in keywords/keyword-groups.json, letting you choose when and how to engage with challenging topics:
| Category | Description |
|---|---|
| Climate and Environment | Terms used in crisis coverage and disaster reporting |
| Economic Policy | Language around economic conflicts, crises, and inequality |
| Education | Terms common in educational controversy and conflict reporting |
| Gun Policy | Terms associated with gun violence and contentious legislation |
| Healthcare and Public Health | Coverage of healthcare crises and system failures |
| Immigration | Terms used in border conflict and immigration controversy |
| International Coverage | Language of international conflicts and diplomatic crises |
| LGBTQ+ | Terms frequently used in contentious legislation and inflammatory coverage |
| Media Personalities | Political commentators and media figures frequently involved in political coverage |
| Military and Defense | Terms related to conflict, weapons, and military threats |
| Political Organizations | Political parties, movements, and advocacy groups |
| Political Rhetoric | Inflammatory phrases and contentious political language |
| Political Violence and Security Threats | Terms related to violence, extremism, and threats |
| Race Relations | Language used in coverage of racial conflict and discrimination |
| Religion | Terms associated with religious conflict and discrimination |
| Reproductive Health | Language around healthcare restrictions and rights conflicts |
| Social Policy | Terms common in coverage of social crises and system failures |
| US Government Institutions | Federal agencies, positions, and institutional terminology |
| US Political Figures - Full Name | Notable figures in contentious political coverage |
| US Political Figures - Single Name | Common references in controversial coverage |
| Violence Against Women | Terms related to violence and abuse |
| Vaccine Policy | Language of vaccine controversy and mandate conflicts |
| World Leaders | International figures in political coverage |
While the default keywords are US-centric, you can easily maintain your own regional keyword list:
-
Create a plain text file with your keywords (one per line), for example:
keyword1 keyword2 keyword3 -
Host this file anywhere on the web (e.g., GitHub Gist, Pastebin)
-
In the extension settings:
- Go to the "Import/Export" tab
- Paste your URL in the "Enter URL to import settings" field
- Click "Import from URL"
-
Enable "Check for new keywords on startup" to automatically update your keywords when you start your browser
- 🔍 Customizable Filtering: Tailor your browsing experience by defining custom keywords and keyword groups.
- 🌐 Platform-Specific Features:
- Reddit: Filter posts based on keywords while keeping subreddits visible. Optionally enable comment thread filtering with granular control over top-level comments vs replies.
- LinkedIn: Smart filtering of posts and articles while preserving professional networking features.
- YouTube: Content filtering across video titles, descriptions, and comments.
- CNN & StackOverflow: Targeted content filtering while maintaining site functionality.
- 🔄 Flexible Matching: Choose between exact and flexible matching to refine content identification.
- 🔒 Domain Control: Enable or disable filtering for specific domains to match your browsing habits.
- 🔗 Updates Section: The keyword list will be updated every few hours with the newest cast of characters and controversies.
- Clone or Download: Get this repository on your local machine.
- Open Chrome Extensions: Navigate to
chrome://extensions/. - Enable Developer Mode: Toggle the switch in the top right corner.
- Load Unpacked: Click and select the directory where this extension is located.
Once installed, the extension works automatically based on your settings. Customize these settings through the options.html page, accessible from the extension's icon in the Chrome toolbar.
In the extension settings under "Elements" tab:
- Reddit Comment Filtering: Toggle "Filter Reddit Comment Threads" to enable/disable comment filtering. When enabled:
- Top-level comments containing filtered keywords are completely hidden
- For reply comments, only the filtered comment is hidden while preserving the thread structure
- Posts are always filtered based on keywords, while subreddits remain visible
To load the extension in Chrome for development:
- Clone or Download the Repository: Get a local copy of the project on your machine.
- Open Chrome Extensions Page: Navigate to
chrome://extensions/in your Chrome browser. - Enable Developer Mode: Toggle the "Developer mode" switch at the top right corner.
- Load Unpacked Extension: Click on the "Load unpacked" button.
- Select the Extension Directory: Browse to the directory where you cloned or downloaded the repository and select it.
- Extension Loaded: The extension should now be loaded into Chrome and ready for testing.
Remember to reload the extension from the extensions page after making changes to see the updates.
- Develop a configurable scale to adjust the sensitivity of the algorithm in identifying containers to hide. This would allow users to fine-tune the content filtering based on their preferences.
Contributions are welcome! If you're a CSS, HTML, or JavaScript expert and can help improve the algorithm for finding containers, I'd love your input. Please feel free to submit a pull request or open an issue with your ideas.
This project is licensed under the MIT License. See the LICENSE file for more details.
I'm a software architect with limited JavaScript expertise. I built this extension primarily using AI tools - specifically Claude.ai's web interface, the Cline VS Code extension, OpenRouter.ai's reseller API service, and Anthropic's direct API. While I can work with JavaScript when needed, I relied heavily on these AI assistants for development.
One way that I optimized costs was to have a Project in Claude solve the problem, then give the output to Cline to do the implementation. I also learned that when debugging, sometimes the session will get in a loop and make worse and worse code. After spending about $1.50 on a problematic bug, I found much better results by having Cline summarize the problem and potential solution, then starting a new session to implement it.
I ended up using OpenRouter when I ran out of Anthropic tokens (their limit is quite low). The total cost was around $300, but that was mainly because I chose the wrong model on OpenRouter (Anthropic/auto?) that didn't have caching, and I was initially feeding it huge files before the refactor.
And really, it's addictive to add new features when it's so easy and fun to use.