Quick Start

Install it once. The first QR scan does the rest.

This page is tuned for the searcher who wants the shortest path from “WeChat AI bridge” to a working message loop. It keeps the instructions factual, compact, and runnable.

View upstream repository Review features

Fastest path

shell install

01curl -sSL https://raw.githubusercontent.com/fastclaw-ai/weclaw/main/install.sh | sh0203weclaw start

Install Options

Start with the method that matches your workstation.

The README exposes shell, Go, and Docker paths. The site turns them into three clear starting points instead of one long block.

Shell

One-line install

The shortest path for local evaluation. Run the installer, then start the bridge and scan the QR code when prompted.

shell

01curl -sSL https://raw.githubusercontent.com/fastclaw-ai/weclaw/main/install.sh | sh

Install with Go

Use the standard Go install flow when you already manage your local tooling that way.

01go install github.com/fastclaw-ai/weclaw@latest

Docker

Run in Docker

Mount the local config directory and start the image when container-first evaluation is a better operational fit.

docker

01docker run -it -v ~/.weclaw:/root/.weclaw ghcr.io/fastclaw-ai/weclaw start

First Run

Three steps from install to routed chat.

The product story is strongest when first-run behavior is explicit: login, auto-detection, and immediate thread routing.

Install

Install the bridge

Use the shell script for the fastest evaluation path, or choose Go or Docker if that better matches your workstation.

install

01curl -sSL https://raw.githubusercontent.com/fastclaw-ai/weclaw/main/install.sh | sh

Scan the QR code once

Run `weclaw start`, authenticate with WeChat, and let the bridge write the local config it will reuse later.

start

01weclaw start

Route

Send the first routed prompt

Talk to the default agent, send a one-off request to a named model, or inspect bridge status without leaving the thread.

chat commands

01/codex write a sorting function02/cc explain this diff03/status

Config reference

~/.weclaw/config.json

01{02  "default_agent": "claude",03  "agents": {04    "claude": {05      "type": "cli",06      "command": "/usr/local/bin/claude",07      "args": ["--dangerously-skip-permissions"]08    },09    "codex": {10      "type": "cli",11      "command": "/usr/local/bin/codex",12      "args": ["--skip-git-repo-check"]13    }14  }15}

Command Examples

Use the same thread, just switch routing intent.

WeClaw only needs a few commands to stay powerful: default agent traffic, explicit aliases, status checks, and proactive send workflows.

hello

Send a message to the current default agent without prefixing an explicit route.

/codex write a sorting function

Route one request to Codex without changing the default agent for the rest of the thread.

/cc explain this diff

Use the Claude alias when you want a shorter, chat-friendly route inside the message box.

/claude

Switch the default agent to Claude and persist that choice in local config.

/status

Check the active default agent and connection state.

weclaw send --to "[email protected]" --text "Hello from weclaw"

Send a proactive outbound message without waiting for the user to speak first.

Config stays explicit

Defaults, aliases, transport modes, and permission flags all live in a single JSON file instead of being scattered across hidden state.

Push outbound messages

Use the `weclaw send` command or the local HTTP endpoint when your automation needs to initiate the thread rather than wait for user input.