WeChat CLI

Query your local WeChat data from the command line.

Chat history · Contacts · Sessions · Favorites · Statistics · Export

中文文档

✨ Highlights

🚀 Zero-config install — npm install -g and you're done, no Python needed
📦 11 commands — sessions, history, search, contacts, members, stats, export, favorites, unread, new-messages, init
🤖 AI-first — JSON output by default, designed for LLM agent tool calls
🔒 Fully local — on-the-fly SQLCipher decryption, data never leaves your machine
📊 Rich analytics — top senders, message type breakdown, 24-hour activity charts
📝 Flexible export — Markdown or plain text, with time range filtering

📥 Installation (For Humans)

AI Agents — skip ahead to "Installation (For AI Agents)" below.

This fork's compatibility fix (0.2.5)

Install or update from this repository to get the macOS WeChat 4.1.13 key extraction fix:

python3 -m pip install --upgrade "git+https://github.com/sunfeihu007/wechat-cli.git@main"
wechat-cli --version

Use a Python virtual environment to avoid system Python restrictions. With Homebrew Python, create one using python3 -m venv ~/.venvs/wechat-cli and activate it using source ~/.venvs/wechat-cli/bin/activate. Confirm version 0.2.5, log in to WeChat on that machine, then run wechat-cli init --force. If administrator access is required, run sudo "$(command -v wechat-cli)" init --force. After re-signing, fully restart WeChat, log in, and repeat the same command.

The npm/PyPI instructions below refer to upstream releases. This fork's fix is distributed through the Git URL above and includes the rebuilt macOS arm64 helper. Other platform extractors are unchanged.

npm (Recommended)

npm install -g @canghe_ai/wechat-cli

Currently ships a macOS arm64 binary. Other platforms can use the pip method below. PRs with additional platform binaries are welcome.

Update to the latest version:

npm update -g @canghe_ai/wechat-cli

pip

pip install wechat-cli

Requires Python >= 3.10.

From Source

git clone https://github.com/sunfeihu007/wechat-cli.git
cd wechat-cli
pip install -e .

📥 Installation (For AI Agents)

Simply paste the following prompt into Claude Code, OpenClaw, or any AI coding agent:

Install or update WeChat CLI from the main branch of https://github.com/sunfeihu007/wechat-cli in a Python virtual environment, confirm version 0.2.5, then refresh keys after WeChat is logged in on this machine.

For example, in Claude Code:

Note: Make sure you have Node.js installed first. You can ask your agent to set it up if needed.

🚀 Quick Start

Step 1 — Initialize

Make sure WeChat is running, then:

# macOS/Linux: may need sudo for memory scanning
sudo wechat-cli init

# Windows: run in a terminal with sufficient privileges
wechat-cli init

This auto-detects your WeChat data directory, extracts encryption keys, and saves config to ~/.wechat-cli/.

On macOS, you'll need to run the sudo command and enter your password:

If you have multiple WeChat accounts logged in locally, you'll be prompted to choose one. Select the account you're currently using (the default is the first one):

If you're unsure which WeChat account is currently active, navigate to the data folder and sort by modification date to find out:

macOS: Grant Full Disk Access to Terminal

Before running init, make sure your terminal app has Full Disk Access:

Open System Settings → Privacy & Security → Full Disk Access
Add your terminal app (e.g. Terminal, iTerm2, or the terminal in your IDE)
Restart the terminal after enabling

Without this permission, the tool cannot access WeChat's data directory and key extraction will fail.

macOS: `task_for_pid failed` Error

On some macOS systems, init may fail with task_for_pid failed even when running with sudo. This is due to macOS security restrictions on process memory access.

WeChat CLI will automatically attempt to fix this by re-signing WeChat with the required entitlement (original entitlements are preserved). Just follow the on-screen instructions:

The tool will re-sign WeChat automatically
Quit WeChat completely (not just minimize)
Reopen WeChat and log in
Run sudo wechat-cli init --force again

If auto re-signing fails, you can do it manually:

# Quit WeChat first, then:
sudo codesign --force --sign - --entitlements /dev/stdin /Applications/WeChat.app <<'EOF'
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
<plist version="1.0">
<dict>
    <key>com.apple.security.get-task-allow</key>
    <true/>
</dict>
</plist>
EOF

Heads up: Re-signing WeChat is safe and will not cause account issues or bans. However, it may affect WeChat's auto-update mechanism. If you notice any feature not working properly, or want to update WeChat to the latest version, simply re-download and reinstall WeChat from the official website — no need to re-run init, your existing config and keys will continue to work.

Step 2 — Use It

wechat-cli sessions                        # Recent chats
wechat-cli history "Alice" --limit 20      # Chat messages
wechat-cli search "deadline" --chat "Team" # Search messages

🤖 Using with AI Agents

WeChat CLI is designed as an AI agent tool. All commands output structured JSON by default.

Claude Code

Add to your project's CLAUDE.md:

## WeChat CLI

You can use `wechat-cli` to query my local WeChat data.

Common commands:
- `wechat-cli sessions --limit 10` — list recent chats
- `wechat-cli history "NAME" --limit 20 --format text` — read chat history
- `wechat-cli search "KEYWORD" --chat "CHAT_NAME"` — search messages
- `wechat-cli contacts --query "NAME"` — search contacts
- `wechat-cli unread` — show unread sessions
- `wechat-cli new-messages` — get messages since last check
- `wechat-cli members "GROUP"` — list group members
- `wechat-cli stats "CHAT" --format text` — chat statistics

Then in conversation you can ask Claude things like:

"Check my unread WeChat messages"
"Search for messages about the project deadline in the Team group"
"Who sent the most messages in the AI group this week?"

OpenClaw / MCP Integration

WeChat CLI works with any AI tool that can execute shell commands:

# Get recent conversations
wechat-cli sessions --limit 5

# Read specific chat
wechat-cli history "Alice" --limit 30 --format text

# Search with filters
wechat-cli search "report" --type file --limit 10

# Monitor for new messages (great for cron/automation)
wechat-cli new-messages --format text

📖 Command Reference

`sessions` — Recent Chats

wechat-cli sessions                        # Last 20 sessions
wechat-cli sessions --limit 10             # Last 10
wechat-cli sessions --format text          # Human-readable

`history` — Chat Messages

wechat-cli history "Alice"                 # Last 50 messages
wechat-cli history "Alice" --limit 100 --offset 50
wechat-cli history "Team" --start-time "2026-04-01" --end-time "2026-04-03"
wechat-cli history "Alice" --type link     # Only links
wechat-cli history "Alice" --format text

Options: --limit, --offset, --start-time, --end-time, --type, --format

`search` — Search Messages

wechat-cli search "hello"                  # Global search
wechat-cli search "hello" --chat "Alice"   # In specific chat
wechat-cli search "meeting" --chat "TeamA" --chat "TeamB"  # Multiple chats
wechat-cli search "report" --type file     # Only files

Options: --chat (repeatable), --start-time, --end-time, --limit, --offset, --type, --format

`contacts` — Contact Search & Details

wechat-cli contacts --query "Li"           # Search contacts
wechat-cli contacts --detail "Alice"       # Contact details
wechat-cli contacts --detail "wxid_xxx"    # By WeChat ID

Returns: nickname, remark, WeChat ID, bio, avatar URL, account type.

`members` — Group Members

wechat-cli members "Team Group"            # All members (JSON)
wechat-cli members "Team Group" --format text

`stats` — Chat Statistics

wechat-cli stats "Team Group"
wechat-cli stats "Alice" --start-time "2026-04-01" --end-time "2026-04-03"
wechat-cli stats "Team Group" --format text

Returns: total messages, type breakdown, top 10 senders, 24-hour activity distribution.

`export` — Export Conversations

wechat-cli export "Alice" --format markdown              # To stdout
wechat-cli export "Alice" --format txt --output chat.txt  # To file
wechat-cli export "Team" --start-time "2026-04-01" --limit 1000

Options: --format markdown|txt, --output, --start-time, --end-time, --limit

`favorites` — WeChat Bookmarks

wechat-cli favorites                       # Recent bookmarks
wechat-cli favorites --type article        # Articles only
wechat-cli favorites --query "machine learning"  # Search

Types: text, image, article, card, video

`unread` — Unread Sessions

wechat-cli unread                          # All unread sessions
wechat-cli unread --limit 10 --format text

`new-messages` — Incremental New Messages

wechat-cli new-messages                    # First: return unread + save state
wechat-cli new-messages                    # Subsequent: only new since last call

State saved at ~/.wechat-cli/last_check.json. Delete to reset.

🔍 Message Type Filter

The --type option (on history and search):

Value	Description
`text`	Text messages
`image`	Images
`voice`	Voice messages
`video`	Videos
`sticker`	Stickers/emojis
`location`	Location shares
`link`	Links and app messages
`file`	File attachments
`call`	Voice/video calls
`system`	System messages

💻 System Requirements

macOS ≥ 26.3.1
WeChat for Mac: legacy hex scanning supports ≤ 4.1.8.100. This branch also reads Config.Cipher, verified on Apple Silicon WeChat 4.1.13 (269631) against all 33 local database keys.

Other WeChat builds may require different layouts. Failed, empty, or invalid scans preserve existing keys. Verified new keys are merged with verified existing keys and saved atomically.

🖥️ Platform Support

Platform	Status	Notes
macOS (Apple Silicon)	✅ Supported	Bundled arm64 binary
macOS (Intel)	✅ Supported	x86_64 binary needed
Windows	✅ Supported	Reads Weixin.exe process memory
Linux	✅ Supported	Reads /proc/pid/mem, requires root

🔧 How It Works

WeChat stores chat data in SQLCipher-encrypted SQLite databases locally. WeChat CLI:

Extracts keys — scans WeChat process memory for encryption keys (init)
Decrypts on-the-fly — transparent page-level AES-256-CBC decryption with caching
Queries locally — all data stays on your machine, no network access

📄 License

Apache License 2.0

⚖️ Disclaimer

This project is a local data query tool for personal use only. Please note:

Read-only — this tool only reads locally stored data, it does not send, modify, or delete any messages
No cloud transmission — all data stays on your local machine, nothing is uploaded to any server
No WeChat ecosystem disruption — this tool does not interfere with WeChat's normal operation, does not automate any actions, and does not violate WeChat's Terms of Service
Use at your own risk — this project is for personal learning and research purposes only. Users are responsible for ensuring compliance with local laws and regulations

🙏 Acknowledgements

This project is built on top of wechat-decrypt, which provides the core WeChat database decryption and data parsing capabilities.

Name		Name	Last commit message	Last commit date
image		image
npm		npm
tests		tests
wechat_cli		wechat_cli
.gitignore		.gitignore
LICENSE		LICENSE
README.md		README.md
README_CN.md		README_CN.md
entry.py		entry.py
pyproject.toml		pyproject.toml

Latest commit

History

Folders and files

Repository files navigation

WeChat CLI

✨ Highlights

📥 Installation (For Humans)

This fork's compatibility fix (0.2.5)

npm (Recommended)

pip

From Source

📥 Installation (For AI Agents)

🚀 Quick Start

Step 1 — Initialize

macOS: Grant Full Disk Access to Terminal

macOS: task_for_pid failed Error

Step 2 — Use It

🤖 Using with AI Agents

Claude Code

OpenClaw / MCP Integration

📖 Command Reference

sessions — Recent Chats

history — Chat Messages

search — Search Messages

contacts — Contact Search & Details

members — Group Members

stats — Chat Statistics

export — Export Conversations

favorites — WeChat Bookmarks

unread — Unread Sessions

new-messages — Incremental New Messages

🔍 Message Type Filter

💻 System Requirements

🖥️ Platform Support

🔧 How It Works

📄 License

⚖️ Disclaimer

🙏 Acknowledgements

⭐ Star History

About

Resources

Stars

Watchers

Forks

Releases

Packages

Contributors

Languages