Famulor

io.famulor由 famulor.io 发布官方发布者v1.0.0更新于 Nov 2, 2026

Manage AI assistants, history, calls, campaigns, contacts, knowledge, messaging, and automations.

已验证Streamable HTTP可网页运行效率与工作流沟通与协作

概览

AI 生成的概览

Famulor 平台的远程 MCP 服务器,让助手管理 AI 语音助手、通话、活动、联系人、知识库与消息渠道。

可以做什么
  • 创建、更新和删除 AI 语音助手,列出语音、语言、模型及 TTS/STT 提供商
  • 查看跨渠道历史与通话记录,发起和管理通话
  • 创建并运行活动,设置重试规则与允许时段
  • 管理联系人、标签、分群、抑制名单与活动成员
  • 创建和维护知识库,包括上传网站抓取的文档
  • 管理电话线路、消息与邮件渠道,以及自动化、Webhook 和工作区设置
适用场景
适合已在 Famulor 工作区中工作、希望用自然语言让助手搭建或调整语音助手、启动和监控活动、查看通话记录,或通过 WhatsApp 等渠道发送消息的场景。
运行要求
使用 app.famulor.io/mcp 远程端点,无需安装或自行托管。需要通过浏览器完成 OAuth 登录,选择工作区并批准权限。服务器对服务器客户端可使用 API 密钥,以 Authorization Bearer 方式传递。Famulor 套餐需包含 Connect AI / MCP 功能,否则端点返回 403。
安装前请注意
工具可创建、更新和删除助手、活动、联系人和知识库,还能拨打电话和发送消息,会直接影响线上工作区数据。访问权限通过 OAuth 工作区权限或 Bearer API 密钥授予,密钥需妥善保管。通话和消息可能按套餐产生费用。

安装

在 SourceWeft 中

  1. 打开 控制台中的 Famulor,将其添加到工作区。
  2. 连接 所需的 凭据。它们会被私密存储,绝不会在公开页面上展示。
  3. 为需要使用其工具的对话启用该服务。

Web executable,通过 Streamable HTTP。 远程服务在工作区中配置后即可从网页运行时运行。

交给智能体安装

把这段提示词粘贴到 Claude Code、Codex、Cursor 等智能体,或 SourceWeft 对话中。智能体会读取该服务器的安装指南,先向你展示发布者、端点和登录方式,经你同意后再把它添加到你的客户端。密钥不会写进任何命令。

阅读 https://sourceweft.com/mcp/io.famulor%2Ffamulor-mcp/install.md,并按其中说明把这个 MCP 服务器接入我的客户端。修改任何配置之前,先告诉我它的发布者、端点和登录方式,等我确认。不要把 API 密钥写进命令或我会提交的文件;问我密钥应从哪里获取。

查看智能体所依据的安装指南

其他 MCP 客户端

把它添加到你客户端的 mcpServers 配置中。 客户端会在首次连接时完成 OAuth 登录。

{
  "mcpServers": {
    "famulor": {
      "type": "http",
      "url": "https://app.famulor.io/mcp"
    }
  }
}

README

[Famulor]

Famulor MCP

Connect Claude, Codex, Cursor, VS Code, and other MCP clients to the complete tenant-safe Famulor platform API.

MCP endpoint · API & MCP settings · Docs

[GitHub] [Famulor] [MCP] [MIT]

[Famulor AI voice agents]


Setup

Add the server in your client, then sign in to Famulor and approve the requested workspace permissions in the browser. Claude and ChatGPT use OAuth; there is no API key to paste into their connector settings.

Server URL: https://app.famulor.io/mcp. It runs inside the Famulor platform, so there is nothing to install or host. Your plan needs the Connect AI / MCP feature; without it the endpoint answers 403.

Claude Code
bash
claude mcp add --transport http famulor https://app.famulor.io/mcp

Add the server first, then complete the browser-based OAuth flow.

Claude Desktop

Settings > Connectors > Add custom connector

https://app.famulor.io/mcp

Copy the URL, paste it into the connector dialog, and sign in when prompted.

Cursor

.cursor/mcp.json

json
{  "mcpServers": {    "famulor": {      "type": "http",      "url": "https://app.famulor.io/mcp"    }  }}

Add the server first, then complete the browser-based OAuth flow.

Windsurf

~/.codeium/windsurf/mcp_config.json

json
{  "mcpServers": {    "famulor": {      "serverUrl": "https://app.famulor.io/mcp"    }  }}

Add the server first, then complete the browser-based OAuth flow.

VS Code

.vscode/mcp.json

json
{  "servers": {    "famulor": {      "type": "http",      "url": "https://app.famulor.io/mcp"    }  }}

Add the server first, then complete the browser-based OAuth flow.

Cline

Cline extension > MCP Servers > Configure

json
{  "mcpServers": {    "famulor": {      "url": "https://app.famulor.io/mcp"    }  }}

Add the server first, then complete the browser-based OAuth flow.

Continue

~/.continue/config.yaml

yaml
mcpServers:  - name: famulor    type: streamable-http    url: "https://app.famulor.io/mcp"

Add the server first, then complete the browser-based OAuth flow.

OpenClaw
bash
openclaw mcp add famulor --url https://app.famulor.io/mcp --transport streamable-http --auth oauthopenclaw mcp login famulor

Or in ~/.openclaw/openclaw.json:

json
{  "mcp": {    "servers": {      "famulor": {        "url": "https://app.famulor.io/mcp",        "transport": "streamable-http",        "auth": "oauth"      }    }  }}

Tip: there is also a ready-made Famulor skill — openclaw skills install famulor-skill.

Hermes Agent
bash
hermes mcp add famulor --url https://app.famulor.io/mcphermes mcp login famulor

Or in ~/.hermes/config.yaml:

yaml
mcp_servers:  famulor:    url: "https://app.famulor.io/mcp"    auth: oauth

Apply in-session with /reload-mcp.

Zed

~/.config/zed/settings.json

json
{  "context_servers": {    "famulor": {      "args": [        "-y",        "mcp-remote@latest",        "https://app.famulor.io/mcp"      ],      "command": "npx"    }  }}

Add the server first, then complete the browser-based OAuth flow.


What you can do

Once connected, your AI assistant discovers the current public tool catalog directly from Famulor. The catalog covers, among other areas:

Assistants — create, update, delete AI voice agents · list voices, languages, LLM/multimodal models, TTS/STT providers · toggle inbound and conversation-ended webhooks

History & calls — inspect cross-channel history and transcripts · make and manage calls

Campaigns — create campaigns, manage retry rules and allowed-hours windows · start, stop and delete

Audience — manage contacts, tags, segments, suppression and campaign membership

Knowledge bases — create knowledge bases · upload website-scraped documents · update / delete

Telephony — phone numbers, number verification, assignments and SIP trunks

Messaging & email — WhatsApp, Instagram, Messenger, Telegram, Slack, Teams, Discord, Google Chat, X, helpdesk and email conversation workflows

WhatsApp — list senders and templates · check 24h session window · send template or freeform messages

Automations & operations — flows, routines, dashboards, integrations, webhooks, booking, workspace settings and account administration

Ask the assistant in plain English, e.g. "Create a German sales assistant using GPT-4.1-mini and the Susi voice", "Start campaign #42 and add these 50 leads", "Show me all calls today where the goal wasn't reached."


How auth works

The endpoint implements MCP OAuth discovery, authorization code with PKCE, dynamic client registration and Client ID Metadata Documents. Your MCP client opens Famulor in a browser, where you sign in on app.famulor.io, choose a workspace and approve the requested permissions.

Server-to-server clients can send an API key instead: create one under Settings > API & MCP on app.famulor.io and pass it as Authorization: Bearer fam_…. Claude and ChatGPT connectors always use OAuth.

Full tool catalog and scopes: docs.famulor.io/mcp/tools-and-scopes


Famulor 1.0 (app.famulor.de)

This MCP server works with workspaces on the Famulor platform at app.famulor.io. Famulor 1.0 (app.famulor.de) stays available; to bring assistants, tools, knowledge, campaigns, automations and your own telephony over, use Migrate from Famulor 1.0.


What is in this repository

The server itself runs inside the Famulor platform. This repository holds what clients and directories need to find it:

FilePurpose
mcp.json, .plugin/plugin.jsonClient and plugin configuration pointing at https://app.famulor.io/mcp
server.jsonEntry in the official MCP Registry (io.famulor/famulor-mcp, remote only)
glama.jsonGlama ownership claim
SUBMISSIONS.md, submissions/, chatgpt-app-submission.jsonDirectory submission material
assets/Logos

The earlier self-hosted server and its npm package famulor-mcp are retired.


License

MIT

来源:README.md,提交 d61aaa8

工具

0
该服务器需要登录,因此 SourceWeft 无法在没有账号的情况下读取其工具,也未收录。已连接账号所能看到的内容可能因账号及其权限而异。

版本历史

1
  1. v1.0.0最新Sep 16, 2026

评分

已满足适用检查项的 75%

该评分衡量此收录条目的完整程度,依据注册中心条目、端点与公开来源计算得出。它不是对服务器本身的评价,也不表示服务器安全或当前可用:可连通性是带日期的检查结果,评分也不包含安装量或用户评价。

  • 近 30 天内可连通必需25 分
  • 已声明认证方式必需10 分
  • 已列出工具15 分

    请让服务器在无凭据时响应 tools/list,或发布可在沙箱中无凭据启动的软件包。

  • 描述必需15 分
  • 图标10 分
  • 文档15 分
  • 许可证10 分

    请在仓库或注册中心条目中声明许可证。