Last Updated: 2026-08-03
1. Introduction
X-Recommender Plus (“we”, “our”, or “us”) is a browser extension for Chrome and Safari for macOS that helps users filter and manage X and Weibo feeds by analyzing post quality using large language models (LLM). This Privacy Policy explains how we collect, use, disclose, and safeguard your information.
2. Data We Collect
2.1 Data You Provide
API Configuration: When you configure an LLM API in the extension, you may provide:
API endpoint URL (e.g., https://api.openai.com/v1)
API key (optional, for cloud LLM services)
Model name
Settings: You can customize various extension settings including:
Scoring prompt (custom instructions for evaluating tweets)
Minimum score threshold
Opacity level for low-quality tweets
Cache duration
2.2 Data We Process
Post Content: The extension analyzes posts in your X or Weibo feed to generate quality scores. This processing happens locally on your device, through Apple system features when selected, or through your configured LLM API.
Post Scores: The extension stores quality scores for posts to improve performance and provide consistent filtering.
2.3 Data We Do NOT Collect
❌ Personal identity information
❌ X/Twitter account credentials
❌ Browsing history (outside of X/Twitter)
❌ Location data
❌ Device information
3. How We Use Your Data
We use your data to:
Provide Core Services: Analyze posts in your X or Weibo feed and apply filters based on quality scores generated by your configured LLM or selected system feature.
Improve Performance: Cache tweet scores to reduce API calls and improve response times.
Sync Settings: Sync your preferences across devices when the browser’s extension sync facility is available and enabled.
4. Data Storage
4.1 Local Storage
Data is stored in the browser extension storage on your device:
API Configuration: Stored in the browser’s sync or local extension storage
Post Scores: Stored in local extension storage with configurable expiration (default: 7 days)
Settings: Stored in sync or local extension storage where supported
4.2 Data Retention
Post scores are automatically deleted after the configured cache period (default: 7 days)
API keys and settings remain until you manually delete them through the extension
4.3 Data Encryption
Data transmitted to LLM APIs is encrypted via HTTPS/TLS
The browser and macOS protect extension storage with platform-level permissions
5. Third-Party Disclosure
5.1 LLM API Providers
When you configure an external LLM API (such as OpenAI, DeepSeek, Ollama, etc.), your configured API credentials and post content may be sent to that provider. We are not responsible for the privacy practices of third-party LLM providers.
Before using any LLM service, please review their privacy policy:
5.2 Local Models
If you use local LLM models (LM Studio, llama.cpp, LocalAI), Apple Intelligence, or Apple Translation, your data is processed through the selected local or system feature and is not sent to an external LLM provider by us. Apple Translation may download language packs through the system when needed.
5.3 We Do Not Sell Data
We do not sell, rent, or share your personal information with third parties for marketing purposes.
6. Security
We take reasonable measures to protect your data:
The browser and Safari extension sandboxes isolate data from other applications
API credentials are stored in encrypted format
No external servers are used for our own data collection
However, no method of electronic storage is 100% secure. While we strive to protect your information, we cannot guarantee absolute security.
7. Your Rights
You have the right to:
Access: View all stored settings and API configurations
Delete: Remove API configurations and cached data through the extension settings
Opt-out: Uninstall the extension to stop all data processing
8. Changes to This Policy
We may update this Privacy Policy from time to time. Any changes will be posted on this page with an updated “Last Updated” date.
9. Contact Us
If you have questions about this Privacy Policy, please contact us through the extension’s support channels.