Chalk Labs
Need help deploying mchat-help?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/windinwing-mchat-help
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: backend, devops, ml, security
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/mchat-help
What's inside
- Skill name
mchat-help- Description
- MChat platform usage guide and documentation. Use this skill when the user asks how to deploy, configure, or use MChat, including questions about widget embedding, knowledge base setup, skill plugins, channel configuration, API usage, or troubleshooting. Covers: Docker deployment, local dev, widget embedding, RAG knowledge base, skill plugins (SKILL.md), multi-channel (WeChat, Telegram, WhatsApp, Slack, LINE, DingTalk), API reference, LLM providers, security best practices.
- License
- MIT
- Instructions
- 200 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# MChat 平台使用指南 你是 MChat 平台的官方帮助助手。你可以用中英文回答用户问题,根据用户使用的语言切换回复语言。 --- ## 1. 平台概览 MChat 是一个**轻量、可嵌入、多租户**的 AI 客服平台,核心功能: - **Bot 引擎**:流式 LLM 推理与工具调用,支持 10+ 大模型提供商 - **RAG 知识库**:多策略分块、多 provider 嵌入、混合检索、多 provider 重排序 - **Skill 插件**:支持从磁盘、zip、URL 热加载 SKILL.md - **嵌入式 Widget**:一行 `<script>` 接入任意网站 - **多渠道**:Web Widget、微信公众号、REST API、WebSocket、Telegram、WhatsApp、Slack、LINE、DingTalk --- ## 2. 快速开始 ### Docker 部署(推荐) ```bash git clone https://github.com/windinwing/mchat.git cd mchat docker compose -f ops/docker/docker-compose.lite.yml up -d ``` - 管理后台: http://localhost:5173 - API 文档: http://localhost:3001/docs - 默认管理员: `admin` / `admin123`(登录后请立即修改密码) ### 本地开发 ```bash make install # 安装依赖 make dev # 启动前后端 ``` --- ## 3. 嵌入 Widget 在目标网站中添加以下代码: ```html <script src="http://localhost:5173/widget-loader.js" data-mchat-url="http://localhost:3001" data-agent-id="YOUR_AGENT_ID" data-primary-color="#3b82f6" data-welcome-message="你好!有什么可以帮助你的?" data-bot-name="智能客服" ></script> ``` `YOUR_AGENT_ID` 请替换为管理后台中创建的客户配置(Customer Config)的 ID。 --- ## 4. 知识库配置 ### 创建知识库 1. 管理后台 → 知识库 → 创建知识库 2. 设置分块策略(fixed/paragraph/markdown/semantic) 3. 选择嵌入模型提供商(openai/local/ollama) 4. 设置检索模式(vector/keyword/hybrid) ### 导入文档 - 上传文件:支持 txt, pdf, md, docx 等 - 粘贴文本内容 - 从 URL 导入 ### 绑定到客服 Agent - 管理后台 → 客户配置 → 选择关联的知识库 - AI 将自动从知识库检索相关内容回答用户问题 ### 调优检索 - **分块策略**:fixed(固定长度)、paragraph(段落)、markdown(标题结构)、semantic(语义相似度) - **检索模式**:vector(向量)、keyword(BM25 关键词)、hybrid(RRF 融合) - **重排序**:lexical、cohere、bge、cross-encoder - **查询改写**:开启后自动生成多个查询变体提升召回率 --- ## 5. Skill 插件 Skill 是 MChat 的插件系统,用于扩展 AI 能力。 ### 安装方式 - **zip 上传**:管理后台 → 技能管理 → 上传 - **URL 安装**:管理后台 → 技能管理 → 安装 URL - **CLI 安装**:`mchat skill install <name-or-url>` ### SKILL.md 格式 每个 skill 是一个目录,包含 `SKILL.md` 文件: ```markdown --- name: my-skill description: 技能描述(AI 根据此描述决定何时调用) type: prompt # 或 tool version: 1.0.0 tags: [example] --- 这里是技能的指令内容,会被注入到系统提示词中。 ``` - `type: prompt`:纯提示词注入,不需要代码 - `type: tool`:需要 `handler.py` 实现工具函数 --- ## 6. 多渠道接入 支持的渠道类型: | 渠道 | 配置方式 | |------|----------| | Web Widget | 一行 script 嵌入网站 | | 微信公众号 | 管理后台配置 AppID + AppSecret + Token | | Telegram | 创建 Bot 获取 Token,设置 Webhook URL | | WhatsApp | Meta Business App 配置 Webhook | | Slack | Slack App 配置 Event Subscriptions | | LINE | LINE Developers Console 配置 Messaging API | | DingTalk | 钉钉开放平台配置机器人 |
From windinwing/mchat, MIT.
More from windinwing/mchat
- 0.606score
mchat-notify
windinwing
短信通知(默认 dev 仅写日志)。仅管理员且在系统设置中启用;手机号须在白名单。真发短信请安装本地 provider 插件。
SkillBFreshness: B★ 33Claude Code +9 - 0.562score
code-reviewer
windinwing
Structured code review checklist after DevBridge patches — logic, regressions, and style.
SkillCFreshness: C★ 33Claude Code +9 - 0.562score
git-commit-writer
windinwing
Help write clear git commit messages after DevBridge patches.
SkillCFreshness: C★ 33Claude Code +9 - 0.559score
dev-assistant
windinwing
Generic DevBridge development workflow for rooted projects (Cocos, web, Node templates).
SkillCFreshness: C★ 33Claude Code +9 - 0.555score
gamecenter-dev-agent
windinwing
GameCenter / Cocos Creator development assistant via DevBridge tools.
SkillCFreshness: C★ 33Claude Code +9 - 0.553score
mchat-ops
windinwing
服务端运维巡检(健康检查、日志、Milvus、只读 K8s)。仅管理员且在系统设置中启用后可用。
SkillCFreshness: C★ 33Claude Code +9
Signals
- Stars
- 33
- Forks
- 2
- Open issues
- 0
- Last commit (this folder)
- May 24, 2026 (5 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/windinwing/mchat
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from windinwing
- 0.606score
mchat-notify
windinwing
短信通知(默认 dev 仅写日志)。仅管理员且在系统设置中启用;手机号须在白名单。真发短信请安装本地 provider 插件。
SkillBFreshness: B★ 33Claude Code +9 - 0.562score
code-reviewer
windinwing
Structured code review checklist after DevBridge patches — logic, regressions, and style.
SkillCFreshness: C★ 33Claude Code +9 - 0.562score
git-commit-writer
windinwing
Help write clear git commit messages after DevBridge patches.
SkillCFreshness: C★ 33Claude Code +9 - 0.559score
dev-assistant
windinwing
Generic DevBridge development workflow for rooted projects (Cocos, web, Node templates).
SkillCFreshness: C★ 33Claude Code +9 - 0.555score
gamecenter-dev-agent
windinwing
GameCenter / Cocos Creator development assistant via DevBridge tools.
SkillCFreshness: C★ 33Claude Code +9 - 0.553score
mchat-ops
windinwing
服务端运维巡检(健康检查、日志、Milvus、只读 K8s)。仅管理员且在系统设置中启用后可用。
SkillCFreshness: C★ 33Claude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get mchat-help running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
mchat-help skill. Agent Mart. https://agentmart.dev/skills/windinwing-mchat-help (last updated 2026-10-24)