wx-favorites-report Review: WeChat Favorites Visualization Tool - One-Click Interactive Reports
Have you ever wondered how much stuff is actually hiding in your WeChat Favorites? All those articles, videos, and links you saved on a whim — how many of them have you never opened again?
wx-favorites-report is an open-source WeChat favorites visualization tool that can decrypt and parse data end-to-end from the encrypted local WeChat Mac database, then generate an interactive HTML report with 7 chart types. Simply put — it turns your WeChat Favorites from a “black box” into a “dashboard.”
1. What is wx-favorites-report
wx-favorites-report is an open-source project by developer zhuyansen, initially released as a Claude Code Skill and later integrated into projects like claude-code-skills-zh.
Its core capabilities can be summarized in three steps:
- Decryption: Extracts keys from the encrypted WeChat Mac client database (SQLCipher) and decrypts it
- Parsing: Converts structured XML data in SQLite into unified JSON
- Visualization: Generates a single-file HTML report with statistical dashboards, monthly trends, type distribution, and more
The entire process can be fully automated — if you use Claude Code, just say “WeChat favorites visualization” and the tool will run through the whole pipeline automatically.
2. Core Features in Detail
2.1 Data Extraction: From Encrypted DB to Plaintext
WeChat Mac 4.x stores favorites data in a SQLCipher-encrypted SQLite database. wx-favorites-report uses frida dynamic injection to hook WeChat’s CCKeyDerivationPBKDF function and capture the key derivation process at runtime.
# Install dependencies
pip3 install frida frida-tools pycryptodome
# Prepare a signed copy of WeChat (bypass Hardened Runtime)
cp -R /Applications/WeChat.app ~/Desktop/WeChat.app
codesign --force --deep --sign - ~/Desktop/WeChat.appDecryption parameters follow SQLCipher 4 standards: AES-256-CBC encryption, HMAC-SHA512 verification, PBKDF2 with 256,000 iterations.
2.2 Statistical Dashboard
The report’s homepage is a data overview panel displaying four core metrics:
- Total Favorites: How many items you’ve saved in total
- Days Spanned: Duration from your first favorite to now
- Daily Average: Average number of favorites per day
- Source Count: How many different sources your favorites come from
2.3 Seven Visualization Charts
| Chart Type | What It Shows | Interaction |
|---|---|---|
| Monthly Trends | Monthly favorite count changes | Line + area chart |
| Type Distribution | Article/video/link/image proportions | Donut chart |
| Source Ranking | Top 15 favorite sources | Horizontal bar chart |
| Activity Heatmap | Day-of-week × hour favorite frequency | Heat matrix |
| Time Distribution | Which time of day you favorite most | Dual bar chart |
| Word Cloud | Keywords extracted from titles and descriptions | Interactive word cloud |
| Tag Cloud | WeChat favorites tag distribution | Clickable tags |
2.4 Favorites Browser
The bottom of the report features a fully functional favorites browser supporting:
- Filter by type (article/video/link/image)
- Filter by tag
- Full-text search
- Multiple sort options
- Paginated browsing
- Click cards to view full content and original links
3. Technical Principles
3.1 WeChat Database Structure
WeChat Mac 4.x stores favorites data in favorite.db, located at:
~/Library/Containers/com.tencent.xinWeChat/Data/Documents/
xwechat_files/<wxid>/db_storage/favorite/favorite.dbUnlike version 3.x which used a dual-table structure (FavItems + FavDataItem), 4.x switched to a single fav_db_item table with XML-formatted content fields.
3.2 The 6 Iterations of Key Extraction
This is the hardest part of the entire project. The developer went through 6 rounds of attempts:
| Round | Method | Result |
|---|---|---|
| 1 | C memory scan for x’hex’ format | Failed: WeChat 4.x doesn’t use this format |
| 2 | Search for raw salt bytes | Failed: False matches on ASCII strings |
| 3 | HMAC brute-force verification | Failed: Filter conditions skipped the real location |
| 4 | Unfiltered brute-force search | Failed: Keys are stored after PBKDF2 derivation |
| 5 | DYLD hook / lldb | Failed: macOS SIP blocked it |
| 6 | frida + CCKeyDerivationPBKDF | Success |
The final solution hooks the system-level PBKDF2 function via frida, capturing the key derivation process when WeChat starts.
3.3 Data Processing Pipeline
Encrypted DB → frida extracts key → PyCryptodome decrypts → SQLite reads
→ XML parsing → Unified JSON → ECharts renders → Single-file HTML4. Installation & Deployment
4.1 Requirements
| Condition | Description |
|---|---|
| OS | macOS (Apple Silicon or Intel) |
| WeChat Version | Mac 4.x, logged in |
| Python | 3.9+ |
| Optional | Claude Code (one-click execution) |
4.2 Manual Installation Steps
Step 1: Install Python dependencies
pip3 install frida frida-tools pycryptodomeStep 2: Prepare a signed copy of WeChat
cp -R /Applications/WeChat.app ~/Desktop/WeChat.app
codesign --force --deep --sign - ~/Desktop/WeChat.appStep 3: Extract keys
killall WeChat 2>/dev/null; sleep 2
# Run frida hook script
# After WeChat starts: log in → open "Favorites" page → wait 60 secondsStep 4: Decrypt and generate report
# Parse data
python3 parse_favorites.py --input favorite_decrypted.db --output data.json
# Generate HTML report
python3 generate_report.py --input data.json --output report.html
# Local preview
python3 -m http.server 8765
# Open http://localhost:8765/report.html in browser4.3 One-Click with Claude Code
If you have Claude Code installed, just say in conversation:
WeChat favorites visualizationClaude will automatically complete key extraction, decryption, parsing, and report generation.
5. Hands-On Tutorial: Generate Your First Favorites Report
5.1 Data Extraction
After running the frida script, WeChat will restart. At this point you need to:
- Log in to WeChat (scan QR code or confirm)
- Click the “Favorites” tab on the left
- Wait 60 seconds for frida to capture the keys
Keys are saved to /tmp/wechat_frida_keys.log, matched by salt to find the enc_key corresponding to favorite.db.
5.2 Database Decryption
Decrypt with Python:
from Crypto.Cipher import AES
import hashlib
# Get enc_key and salt from frida log
enc_key = bytes.fromhex("your_enc_key_here")
# SQLCipher 4 parameters
kdf_iter = 256000
page_size = 4096
reserve_size = 80 # IV(16) + HMAC(64)
# Decrypt page by page...5.3 Report Generation & Customization
python3 generate_report.py --input data.json --output report.htmlThe generated report is a single-file HTML with all CSS and JavaScript inlined. You can double-click to open it directly, or upload it to any static hosting service.
6. Reading the Report: How to Analyze Your Favorites Data
6.1 Discover Usage Patterns
Through the activity heatmap, you can discover:
- Do you favorite more on weekdays or weekends?
- What time of day is your peak favoriting period?
- Is there a “late-night phone scrolling and mass favoriting” pattern?
6.2 Content Preference Analysis
The type distribution chart reveals your favorites preferences:
- If articles make up 80%+, you’re a “deep reading” user
- If videos dominate, you prefer “fragmented learning”
- If links dominate, it’s likely more “tool bookmarking”
6.3 Source Quality Assessment
The Top 15 Source Ranking helps you identify:
- Which official accounts/websites do you favorite most?
- Is the content quality from these sources consistently stable?
- Are there sources worth unfollowing or adding to an RSS feed instead?
7. Comparison with Other WeChat Data Tools
| Tool | Data Type | Platform | Visualization | Difficulty |
|---|---|---|---|---|
| wx-favorites-report | Favorites | Mac | ⭐⭐⭐⭐⭐ | Medium |
| WeChatMsg | Chat history | Windows | ⭐⭐⭐ | Medium |
| PyWxDump | Chat history + contacts | Windows | ⭐⭐ | High |
| WeChat built-in favorites search | Favorites | All platforms | ⭐ | Low |
wx-favorites-report’s unique advantages:
- Focused on the favorites scenario: Not as large and complex as chat history tools
- High visualization quality: 7 professional chart types, far beyond WeChat’s built-in list search
- Single-file output: View without installing additional software
- Claude Code integration: Complete the entire workflow with one sentence
If you’re also interested in WeChat chat history, check out our previous article PyWxDump Review.
8. Use Cases
8.1 Personal Data Management
Regularly generating favorites reports helps you:
- Clean up outdated or useless favorites
- Discover forgotten quality content
- Build a material list for your personal knowledge base
8.2 Content Organization
If you’re a content creator, favorites reports can:
- Analyze the fields and topics you follow
- Discover patterns in where your content inspiration comes from
- Provide data support for topic selection
8.3 Usage Habit Analysis
Reflect on your information consumption patterns from the data:
- Is it “save and forget”?
- Is your favoriting behavior related to anxiety?
- How can you build a more effective information management system?
9. Limitations & Notes
9.1 Platform Limitations
- macOS only: Key extraction relies on frida and the macOS WeChat client
- WeChat version requirement: Currently adapted for Mac 4.x; 3.x has a different table structure
- WeChat must be logged in: You need an actively used WeChat account
9.2 Data Privacy
- All data processing is done locally — nothing is uploaded to any server
- The report HTML file contains your favorites data — do not share it publicly
- frida hooks the WeChat process while running; it’s recommended to close the signed copy after completion
9.3 Known Limitations
- Original content for images/videos/files is stored on WeChat CDN (encrypted format) and cannot be previewed in the report
- Article thumbnails come from WeChat official account CDN; internet access is needed to view the report
- The desktop copy may need to be re-signed after each WeChat update
10. FAQ
Q1: Is wx-favorites-report safe? Will it leak my WeChat data?
All data processing is done entirely locally — nothing is uploaded to any server. frida only hooks the WeChat process locally to extract keys, and the generated report is a purely local HTML file. However, please keep the generated report file safe and do not share it publicly.
Q2: Does it support Windows or mobile WeChat?
Currently it only supports the WeChat Mac 4.x client on macOS. The encryption mechanisms on Windows and mobile are different and not yet supported. If you need a WeChat data tool for Windows, check out PyWxDump.
Q3: Will it still work after WeChat updates?
Major WeChat version updates may change the database structure or encryption method, in which case you’ll need to wait for the tool to be adapted. Minor updates usually don’t affect usage, but the desktop copy needs to be re-signed after each update.
Q4: Can I share the generated report?
The report is a single-file HTML that can be viewed offline. However, since it contains your personal favorites data, it’s recommended for personal use only — do not upload it to public websites or share it with others.
Q5: Can I use it without Claude Code?
Yes. Claude Code simply provides the convenience of one-click execution; you can complete the entire workflow manually by following the documentation steps. The core scripts (parse_favorites.py and generate_report.py) are standalone Python programs.
11. Summary
wx-favorites-report is the most complete WeChat favorites visualization solution I’ve seen. It solves the pain point of “WeChat favorites data cannot be exported and analyzed,” with a fairly mature full-chain automation from encrypted database to interactive report.
Pros:
- Full-chain automation, from decryption to report with one click
- 7 visualization chart types with rich data insights
- Single-file HTML output, zero-dependency viewing
- Claude Code integration lowers the barrier to entry
Cons:
- macOS only — Windows/mobile users cannot use it
- The key extraction process has some technical barriers
- Image/video content cannot be previewed in the report
If you’re a “heavy favoriting user” who wants to figure out what you’ve actually saved and what your favoriting habits look like, wx-favorites-report is worth a try.
Hope this blog post was helpful! If you’re interested in WeChat data management, check out our AI Tools Review Collection to learn more about practical data analysis and visualization tools.
