multi-layered browser protection against phishing, malicious javascript, and compromised extensions.
built in response to recent attacks where trusted browser extensions went rogue and executed malicious javascript, compromising millions of users.
this extension is currently in development and NOT ready for production use.
- known bugs and issues exist
- active development and testing in progress
- contributions, bug reports, and PRs are welcome
- planning to publish on chrome web store after stability improvements
do NOT use this for any public-facing product or in production environments.
- tensorflow.js neural network (4-layer deep nn)
- 20-feature extraction from urls
- 95%+ accuracy on validation datasets
- real-time inference (<10ms per url)
- trained on phishtank and custom datasets
- statistical anomaly detection with z-scores
- learns "normal" behavior for each website (first 5 visits)
- detects suspicious changes: new scripts, network requests, api usage
- catches compromised legitimate websites
- blocks dangerous patterns:
eval(),Function(), innerHTML exploits - catches crypto miners (coinhive, cryptoloot, etc.)
- prevents keyloggers and data exfiltration
- pattern-based detection with severity scoring
- monitors OTHER installed extensions
- detects permission changes and version updates
- flags suspicious host permission additions
- risk scoring for extension behavior
- 100% local processing (zero external communication)
- no data collection or cloud services
- no telemetry or tracking
- all analysis happens in your browser
multi-layer protection:
- layer 1 (pre-load): ml classifies urls, blocks phishing before page loads
- layer 2 (load-time): heuristic analysis catches structural anomalies
- layer 3 (runtime): javascript blocking prevents malicious code execution
- layer 4 (post-load): statistical monitoring detects behavioral changes
this is a plasmo extension project bootstrapped with plasmo init.
- node.js 16+
- npm or pnpm
# clone the repository
git clone https://github.com/subhanUmer/SecureWeb-Extension.git
cd SecureWeb-Extension
# install dependencies
npm install
# or
pnpm installfirst, run the development server:
npm run dev
# or
pnpm devopen your browser and load the appropriate development build. for chrome with manifest v3, use: build/chrome-mv3-dev.
you can start editing the popup by modifying src/popup/index.tsx. it should auto-update as you make changes.
for further guidance, visit plasmo documentation
run the following:
npm run build
# or
pnpm buildthis creates a production bundle in build/chrome-mv3-prod/, ready to be loaded as an unpacked extension or zipped for distribution.
npm run prepare-datasetnpm run train-modeltrained model will be saved to assets/ml-models/threat-classifier/
contributions are highly welcome. this project has known issues and bugs that need fixing.
- fork the repository
- create a feature branch (
git checkout -b feature/your-feature) - commit your changes (
git commit -m 'add some feature') - push to the branch (
git push origin feature/your-feature) - open a pull request
- bug fixes and stability improvements
- performance optimizations
- ui/ux enhancements
- additional ml features and datasets
- test coverage
- documentation improvements
special thanks to:
- ahmed bhatti - contributor
- shameer hassan - contributor
this project is open source. check the repository for license details.
this extension is provided "as-is" without warranty of any kind. use at your own risk. not intended for production environments or public-facing products at this stage.
for questions, suggestions, or collaboration:
- github issues: report bugs or request features
- contributions: pull requests welcome
stay safe out there.