wacrew
Open-source WhatsApp AI automation: inbox sorted by wait time, replies in your own voice with human-paced delivery, client CRM, live 2D office and an MCP server for Claude Code and Codex.
Install / Use
claude mcp add sumioshi -- npx -y github:sumioshi/wacrewIf the server publishes to npm under a different name, use that package instead — check the repo README.
MCP Server
Model Context Protocol server
Quality Score
Category
AutomationSupported Platforms
Our assessment of wacrew
wacrew scores 84/100 on our quality scale, 1988th of 2,758 Automation skills we index.
Its MCP Server is 16 KB long, well organised into 11 sections with 9 code examples: a thorough specification that gives an agent plenty to work with.
It has 10 GitHub stars, so there is little community track record yet; judge it on its content.
Maintenance, license and trust
- The repository was last updated 3 days ago, so wacrew is actively maintained.
- It is released under the Apache-2.0 license, a permissive license that allows use, modification and commercial use with attribution.
- Its trust signals score 97/100, with no cautions. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.
wacrew compared with similar skills
All 4 of these similar skills score higher than wacrew; compare them before choosing.
| Skill | Score | Stars | Updated | Format |
|---|---|---|---|---|
| wacrew (this skill)by sumioshi | 84 | 10 | 3d ago | MCP Server |
| Agent-Reachby Panniantong | 100 | 114.0k | today | CLAUDE.md |
| hyperframesby heygen-com | 100 | 88.2k | today | CLAUDE.md |
| rufloby ruvnet | 100 | 82.3k | today | CLAUDE.md |
| headroomby headroomlabs-ai | 100 | 80.5k | today | CLAUDE.md |
Frequently asked questions
- How do I install wacrew?
- Run
claude mcp add sumioshi -- npx -y github:sumioshi/wacrew. The install tabs above show the steps for each supported agent. - Which AI agents does wacrew work with?
- It is written for Claude Code, Claude Desktop and OpenAI Codex, as a MCP Server file. Other agents that read the same format can often use it too.
- Is wacrew safe to use?
- It is Apache-2.0-licensed and scores 97/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
- Is wacrew still maintained?
- The repository was last updated 3 days ago, so wacrew is actively maintained.
Skill content
View source on GitHubWhy
If you sell or deliver work over WhatsApp, you know the scene: forty client groups, a four-minute voice note in the middle of a thread, the bug screenshot lost in the scroll, and the client who wrote three days ago that nobody saw.
A chatbot does not fix that. Canned answers read like a robot, clients notice, and the relationship cools down. WACrew works differently:
- Nobody waits. Every group and chat goes into one queue, oldest wait first, with voice notes already transcribed.
- It sounds like you. Replies follow your punctuation, openings and rhythm, measured on your own messages, and pass a review that blocks robotic tells before they leave.
- It moves like a person. Read receipt, typing, pauses between bubbles, and it stops if the client writes again or you answer from your phone.
- You stay in charge. Confirm mode is the default: nothing goes out without your OK. If a client asks whether it is AI, it does not lie; it stops and hands the chat back to you.

The engine that makes it read like a person:
| 2 | 25+ | 60 | ~30 s | |:---:|:---:|:---:|:---:| | writing skills: your voice and client conduct | code rules that block robot tells | history messages read before replying | median until typing starts in a live chat |
What it does
Nobody waits. Every group and direct chat lands in one queue sorted by waiting time. A client's group and their private chats show up together, back-to-back messages collapse into one block, and voice notes arrive transcribed.
Writes like you, delivers like a person. The AI reads the history and writes with your punctuation, openings and rhythm, measured on your own past messages. Before anything leaves, a deterministic review blocks robot clichés, repeated apologies and duplicate bubbles (the generic part of it is also a standalone library, botsmell). Delivery comes with read receipt, typing indicator and pauses, and it stops on its own if the client writes again or you answer from your phone.
You stay in control. In Confirm mode (the default) nothing is sent without your OK. Autonomous mode is turned on per chat. Every message records who wrote it: you on the phone, you on the panel, a teammate, or the AI.
| | |
|---|---|
|
Today. Who is waiting and for how long, what the AI is doing, signals and today's agenda. |
2D office. One desk per project. When a client writes, the desk turns red right away, with sound. |
|
Dashboard. Volume, first-reply time, who answered, busiest clients and deliveries per month. |
Clients. Table and board views with contract, deliveries and waiting time per client. |
|
Agent bridge (optional). If you use Orca, the client drawer shows the coding agent working on that project's repo and lets you send it commands. |
Calendar. Your Google Calendar inside the panel, editable by you or by the AI. |
More screens in docs/screenshots/: org chart, team access, notifications, project links, settings and the command palette.
How it works
┌────────────────────────────┐
WhatsApp ─► collector ─►│ data/<chat>/ │◄─► web panel (Next.js, :3000)
(Baileys) (daemon) │ messages.jsonl + media │
│ transcripts, notes, state │◄─► MCP server (your coding agent)
└─────────────┬──────────────┘
transcriber (MLX on Apple Silicon, or an API)
- Collector (
src/): a Baileys daemon (WhatsApp Web multi-device, no browser). Baileys is isolated behind ports and adapters insrc/whatsapp/. It writes every message, reaction and media file todata/<chat>/and runs the delivery queue that sends replies at a human pace. - Panel (
web/): Next.js 16 + React 19. Inbox, conversation view, clients, dashboard, calendar, team, 2D office, settings. Live updates over SSE. - MCP server (
web/mcp/server.ts): 37 tools over the same data, for Claude Code, Codex, Antigravity, Gemini CLI, Qwen Code or any MCP client. - Transcriber (
transcriber/): Whisper large-v3 via MLX, loaded on demand and released when idle. Off a Mac, setTRANSCRIBE_BACKEND=openrouter. - Local or cloud. Run everything on your machine, or deploy the collector and panel 24/7 with the included
Dockerfile(arailway.jsonis provided). The local MCP can read straight from the cloud panel (WAC_MCP_REMOTE=1), downloading media on demand.
Quick start
Requirements: Node.js 20+, ffmpeg, and a WhatsApp account to pair. Local transcription needs Apple Silicon (uv tool install mlx-whisper).
git clone https://github.com/sumioshi/wacrew && cd wacrew
npm install && npm --prefix web install
cp .mcp.json.example .mcp.json
npm run dev # collector + panel + transcriber + notifier
Open http://localhost:3000/config, scan the QR code (WhatsApp › Linked devices › Link a device), pick the groups to follow and mark who is on your team. The session is saved in auth/ and reconnects on its own.
Run it 24/7 on a VPS. WACrew only records what arrives while it is running: if your computer is off, messages from that period never reach the history or the queue. The repo ships a docker-compose.yml with a persistent volume for the WhatsApp session and data:
cp .env.example .env # set PANEL_USER and PANEL_PASS
docker compose up -d
<a href="https://www.hostinger.com/vps-hosting?REFERRALCODE=ORISHODIRBUH"><img alt="Deploy on Hostinger" src="docs/assets/deploy-hostinger-en.svg" height="32"></a> <a href="https://railway.com/deploy/wacrew?referralCode=8ZIi5q&utm_medium=integration&utm_source=template&utm_campaign=generic"><img alt="Deploy on Railway" src="https://railway.com/button.svg" height="32"></a>
<sub>Referral links: you get a discount or credits, and the project earns a commission that helps keep it maintained. On Railway the template sets up the volume, the login and a generated password for you.</sub>
Other clouds. Run the prebuilt image ghcr.io/sumioshi/wacrew or build the Dockerfile on any host with a volume at /data. Set PANEL_USER and PANEL_PASS to protect the panel with login; without them the panel is open, which is only fine on localhost.
MCP tools
The server is registered by .mcp.json (project scope). Claude Code and Qwen Code read it and ask for approval on first use.
From npm, without cloning. With the panel running on a VPS, add the MCP to any client with wacrew-mcp:
claude mcp add wacrew -e WAC_CLOUD_URL=https://your-panel.example.com \
-e WAC_CLOUD_USER=admin -e WAC_CLOUD_PASS=change-me -- npx -y wacrew-mcp
Antigravity plugin. The repo is also an Antigravity plugin, with the MCP and the skills (plantao, autonomo, link-whatsapp, comunicacao-cliente):
agy plugin install https://github.com/sumioshi/wacrew
It installs turned off. Open /plugin (or Customizations in the editor), turn on WACrew and fill in the panel URL, login and password; other accounts go in the URL field after a space, as below. To add only the MCP, flags go before the name:
agy mcp add -e WAC_CLOUD_URL=https://your-panel.example.com \
-e WAC_CLOUD_USER=admin -e WAC_CLOUD_PASS=change-me wacrew npx -y wacrew-mcp
More than one WhatsApp. One server can talk to several panels. Add the others in WAC_CONTAS as name=https://login:password@address, separated by spaces (or put them after the main address in WAC_CLOUD_URL). Every tool then takes an optional conta parameter, and the agent picks the account per call. WAC_CONTA_NOME names the default one:
claude mcp add wacrew -e WAC_CLOUD_URL=https://work.example.com \
-e WAC_CLOUD_USER=admin -e WAC_CLOUD_PASS=change-me -e WAC_CONTA_NOME=work \
-e WAC_CONTAS="personal=https://admin:other-pass@personal.example.com" -- npx -y wacrew-mcp
| Area | Tools |
|---|---|
| Read | listar_grupos, ler_mensagens, buscar, listar_contatos, caixa_de_entrada, novidades, resumo_do_dia, ler_notas, estado_triagem |
| Media | transcrever, ver_imagem, ver_video, assistir_video, ler_documento |
| Reply | preparar_resposta,
Truncated for display — read the full file on GitHub.
Related Skills
Agent-Reach
114.0kGive your AI agent eyes to see the entire internet. Read & search Twitter, Reddit, YouTube, GitHub, Bilibili, XiaoHongShu — one CLI, zero API fees.
hyperframes
88.2kAI video you can actually edit. HyperFrames is a free, open-source AI video editor. Describe any video and get a real project you can change down to the frame.
ruflo
82.3k🌊 The original agent harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. Features adaptive memory, self-learning intelligence, federation, vector RAG integration, and native Claude Code / Codex / Hermes and many more Integrated
headroom
80.5kCompress tool outputs, logs, files, and RAG chunks before they reach the LLM. 20% fewer tokens for coding agents, 60-95% fewer tokens for JSON, same answers. Library, proxy, MCP server.
Languages
Trust signals
From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.
