MChat 是 AI Skill Hub 本期精选AI工具之一。综合评分 7.5 分,整体质量较高。我们推荐使用将其纳入你的 AI 工具库,帮助提升工作效率。
MChat是一款轻量级、可嵌入的多租户AI客户服务平台,提供高效的客户服务解决方案。
MChat 是一款基于 Python 开发的开源工具,专注于 tag1、tag2、tag3 等核心功能。作为 GitHub 开源项目,它拥有活跃的社区支持和持续的版本迭代,代码完全透明可审计,支持本地部署以保护数据隐私。无论是个人使用还是集成到企业工作流,都能提供稳定可靠的解决方案。
MChat是一款轻量级、可嵌入的多租户AI客户服务平台,提供高效的客户服务解决方案。
MChat 是一款基于 Python 开发的开源工具,专注于 tag1、tag2、tag3 等核心功能。作为 GitHub 开源项目,它拥有活跃的社区支持和持续的版本迭代,代码完全透明可审计,支持本地部署以保护数据隐私。无论是个人使用还是集成到企业工作流,都能提供稳定可靠的解决方案。
# 方式一:pip 安装(推荐)
pip install mchat
# 方式二:虚拟环境安装(推荐生产环境)
python -m venv .venv
source .venv/bin/activate # Windows: .venv\Scripts\activate
pip install mchat
# 方式三:从源码安装(获取最新功能)
git clone https://github.com/windinwing/mchat
cd mchat
pip install -e .
# 验证安装
python -c "import mchat; print('安装成功')"
# 命令行使用
mchat --help
# 基本用法
mchat input_file -o output_file
# Python 代码中调用
import mchat
# 示例
result = mchat.process("input")
print(result)
# mchat 配置文件示例(config.yml) app: name: "mchat" debug: false log_level: "INFO" # 运行时指定配置文件 mchat --config config.yml # 或通过环境变量配置 export MCHAT_API_KEY="your-key" export MCHAT_OUTPUT_DIR="./output"
MChat is a lightweight, embeddable, multi-tenant vertical RAG platform. It combines a streaming Bot engine, production-grade RAG knowledge base, hot-reload Skill plugins, visual Workflow orchestration (Beta), and a one-line embeddable Widget — with 10+ LLM providers and multi-channel delivery (Web Widget, WebSocket, REST, WeChat Official Account, and more).
Built-in AI customer service works out of the box. Extend the same stack into vertical channels — patent search, medical, legal, and other domain RAG packages with dedicated knowledge bases, skill packs, and tuned retrieval — embed anywhere with a single <script> tag.
SKILL.md from disk/zip/URL, OpenClaw-compatible formats; premium packs as vertical add-ons<script> tag for branded vertical RAG chat on any websitemake docker-up-lite full stack```bash git clone https://github.com/windinwing/mchat.git cd mchat
make docker-up-lite
git pull make docker-up-lite
**First clone**:
bash git clone https://github.com/windinwing/mchat.git cd mchat make setup && make dev # or: make docker-up-lite
**Docker / MySQL commands**:
| Command | Description |
|---------|-------------|
| `make docker-up-lite` | Create `.env`, fix MySQL password, build & start |
| `make docker-down-lite` | Stop stack (keep volumes) |
| `make docker-logs-lite` | Tail logs |
| `make db-mysql-dev` | MySQL only (for `make dev`, default host port **3307**) |
| `make db-docker-reset-lite` | Wipe MySQL volume (password mismatch) |
| `MCHAT_RESET_FORCE=1 make reset-fresh` | Full wipe for clean re-test |
**MySQL (single lite config)**:
- Credentials: `mchat` / `mchat123`, database `mchat`, host port **3307** (see `ops/docker/.env`).
- `make setup` syncs `DATABASE_URL` in `src/backend/.env` automatically.
- Existing MySQL: `MCHAT_SETUP_MYSQL=0 make setup` and set `DATABASE_URL` yourself.
- `Access denied for user 'mchat'`: `make db-docker-reset-lite && make setup`.
- `make dev` stops Docker frontend/backend if they occupy port 5173 (MySQL is kept).
**Step by step** (without full `make setup`):
bash make install make db-mysql-dev make dev bash make test make lint source scripts/env.sh && mchat run
| File | Services | Use case |
|---|---|---|
docker-compose.lite.yml | MySQL + Backend + Frontend | **Default** (make setup / make docker-up-lite) |
docker-compose.yml | + Milvus, etcd, MinIO, Redis | Full RAG |
docker-compose.prod.yml | + Nginx HTTPS | Production |
| Group | Path | Description |
|---|---|---|
| Chat | /api/chat/* | Conversations and messages |
| Agents | /api/agents/* | AI configuration |
| Knowledge | /api/knowledge/* | Documents and retrieval |
| Widget | /api/widget/* | Embedded chat API |
| Skills | /api/skills/* | Skill management |
| Workflows | /api/workflows/* | Workflow CRUD, runs, templates (Beta) |
| Channels | /api/channels/* | WeChat and other channels |
| Channel Templates | /api/channels/templates/* | One-click vertical channel creation |
| Speech | /api/speech/* | Voice transcription |
| Auth | /api/auth/* | Login / JWT |
| WebSocket | /ws | Real-time streaming |
| Health | /api/health | Service status |
See docs/api.md or /docs (Swagger) after startup.
MChat Workflow chains multiple Skills into reusable pipelines — not just one-shot chat, but multi-step automation with parallel branches, conditional routing, human approval, loop iteration, node groups, and structured outputs (reports, exports, alerts).
| Concept | Role |
|---|---|
| **Skill** | Smallest unit — a tool, function, or webhook (patent-search, custom packs, etc.) |
| **Workflow** | Ordered steps or a **graph** (graph_json) that wires Skills together |
| **Trigger** | **Manual** (admin run-once), **Schedule** (cron via Skill Schedules + worker), or **Channel** (message rules on WeChat / Telegram / Web) |
ComfyUI-style visual graph editor (Admin → Workflows → Edit Graph):
start · skill · condition (10 operators) · approval · merge · batch (loop) · group (visual) · endCmd/G to create a colored, collapsible, resizable group frame${input.keyword}, ${nodes.<id>.result.xxx}, ${item} templates per node; type-preserving variable linking${item} context, concurrent execution==, !=, >, <, >=, <=, contains, not_contains, startswith, endswith); both left and right support template variablesTemplates: built-in flows such as Patent Multi-Dimension Report (search → parallel analysis → merge → chart/Excel/Word/PPT). Browse templates in the gallery with mini graph preview, filter by category, and apply with one click. Save any workflow as a reusable template.
Example: one patent-search skill can appear in many nodes with different payloads — command: search for retrieval, command: analysis + dimension: applicant|ipc|… for parallel analytics, then patent-report for export.
Getting started
make setup && make dev (or make docker-up-lite)patent-search, patent-report) under Admin → Skillsmake dev-worker or set WORKER_ENABLED=true in .envComplete guide: Workflow Orchestrator · 中文版 · Product tour · API: docs/api.en.md#workflows-beta
MChat 是一个轻量级、可嵌入的多租户垂直领域 RAG(检索增强生成)平台。它专为构建特定领域的 AI 应用而设计,支持多租户隔离,能够轻松集成到现有业务系统中,为用户提供专业且精准的知识问答服务。
MChat 提供强大的 Bot engine,支持 OpenAI、Anthropic、Google、DeepSeek、Ollama 及 Groq 等主流 LLM 的流式推理与 Tool calling。通过 Skill plugins 功能,支持从本地、Zip 或 URL 热加载 SKILL.md 插件包(兼容 OpenClaw 格式)。此外,内置的 RAG knowledge base 支持多策略分块与多供应商 Embedding(包括 OpenAI 及本地 Ollama),实现高效的混合检索。
推荐使用 Docker 进行部署。您可以克隆仓库后,根据需求选择不同的 Docker Compose 配置文件:`docker-compose.lite.yml` 适用于开发或轻量级场景(仅包含 MySQL、Backend 与 Frontend);`docker-compose.yml` 包含完整的 RAG 组件(Milvus, etcd, MinIO, Redis);`docker-compose.prod.yml` 适配生产环境(含 Nginx HTTPS);而 `docker-compose.dev.yml` 则支持热重载,方便本地开发调试。
项目已提供 Quick start 快速入门指南,帮助开发者通过简单的命令快速启动服务并进行初步体验。
MChat 支持针对垂直渠道(Agent)进行精细化配置。开发者可以通过管理界面对 AI Agent 进行定制化设置,以满足不同业务场景下的交互需求。
MChat 提供完善的 RESTful API 接口,文档地址为 http://localhost:3001/docs。API 分为五个核心分组:Chat(对话与消息管理)、Agents(AI 配置管理)、Knowledge(文档与检索管理)、Widget(嵌入式聊天组件 API)以及 Skills(技能插件管理)。
MChat是一款轻量级的AI客户服务平台,提供高效的客户服务解决方案,适用于需要高效客户服务的企业和组织。
AI Skill Hub 为第三方内容聚合平台,本页面信息基于公开数据整理,不对工具功能和质量作任何法律背书。
建议在沙箱或测试环境中充分验证后,再部署至生产环境,并做好必要的安全评估。
✅ MIT 协议 — 最宽松的开源协议之一,可自由商用、修改、分发,仅需保留版权声明。
经综合评估,MChat 在AI工具赛道中表现稳健,质量良好。如果你已有明确的使用需求,可以直接上手体验;如果还在评估阶段,建议对比同类工具后再做决策。
| 原始名称 | mchat |
| 原始描述 | 开源AI工具:MChat is a lightweight, embeddable, multi-tenant AI customer service platform. I。⭐4 · Python |
| Topics | tag1tag2tag3 |
| GitHub | https://github.com/windinwing/mchat |
| License | MIT |
| 语言 | Python |
收录时间:2026-05-24 · 更新时间:2026-05-30 · License:MIT · AI Skill Hub 不对第三方内容的准确性作法律背书。