Reply.io

Name: Reply.io
Author: reply-team

作者：reply-team版本 1.0.0i★ –i在云端运行

注册名称io.github.reply-io/reply-mcp

安装

概览

作者提供的说明（英文原文）

Manage sequences, contacts, and outreach statistics on Reply.io's sales engagement platform.

PopNx MCP 的三大支柱

这三项永远不会合并成一个分数：意见、测量和人气是不同的东西。

用户评价i暂无评价评价功能将在后续阶段与账户功能一起开放
我们的基础检查i1 项中通过 0 项依据公开信息检查；尚未扫描代码
使用情况i尚未统计★ –

它能帮你做什么

我们还没有为它写通俗的介绍。下面是我们从注册中心读取到的事实。

连接方式: 在云端运行
传输方式: Streamable HTTP
分类: 其他
你需要填写：X-API-Key

安装

选择你使用的 AI 应用，然后复制设置。

安装之前

基础检查：1 项中通过 0 项
在作者的服务器上运行：你向它提出的内容会经过他们的服务器。
需要你自己填写： X-API-Key

标为机密的值是你账户的钥匙。千万不要分享给别人，也不要粘贴到聊天里。

你需要

在云端运行无需安装任何东西，通过互联网连接

你用的是哪个应用？

ClaudeClaude CodeCursorVS CodeChatGPT

在上方选择你的应用。我们会记住它，下次按钮会显示“安装到 …”。

这个地址需要在请求的 header 中放入密钥，而添加连接器的菜单无法设置。请改用 Claude Code、Cursor 或 VS Code。

在终端中运行这条命令

claude mcp add --transport http reply https://mcp.reply.io --header 'X-API-Key: your-value'

运行前请先填写空着的值。

在 Cursor 中打开

或者自己复制设置

粘贴到项目中的 .cursor/mcp.json，或 ~/.cursor/mcp.json

{
  "mcpServers": {
    "reply": {
      "url": "https://mcp.reply.io",
      "headers": {
        "X-API-Key": "your-value"
      }
    }
  }
}

浏览器会先询问是否打开应用，然后应用会显示设置并请你确认。密钥的值会留空，由你自己填写。

在 VS Code 中打开

或者自己复制设置

粘贴到项目中的 .vscode/mcp.json

{
  "servers": {
    "reply": {
      "type": "http",
      "url": "https://mcp.reply.io",
      "headers": {
        "X-API-Key": "your-value"
      }
    }
  }
}

浏览器会先询问是否打开应用，然后应用会显示设置并请你确认。密钥的值会留空，由你自己填写。

这个地址需要在请求的 header 中放入密钥，而添加连接器的菜单无法设置。请改用 Claude Code、Cursor 或 VS Code。

你的选择通过 Cookie 保存在这台设备上，无需账户。

需要你填写的值

X-API-Key 机密必填 HTTP header
Reply.io API key from your account settings

这些设置是根据作者在注册中心的条目生成的。如果 README 的写法不同，请以 README 为准。

它能做什么

我们还没有读取这个服务器的工具列表；我们会轮流检查各个端点。

基础检查i

我们检查的是任何人都能看到的信息；我们没有阅读代码。不适用的检查项不计入总数。 · 检查于 2026年11月7日

! 注意
作者与仓库一致
发布者（reply-io）不是仓库所有者（reply-team）；可能是其他人代为上架的
… 尚未检查
仍在维护
我们还没有从 GitHub 读取这个仓库
… 尚未检查
有许可证
我们还没有从 GitHub 读取这个仓库
… 尚未检查
端点有响应
尚未检查；我们会轮流检查各个端点
… 尚未检查
使用 OAuth 登录
尚未检查；我们会轮流检查各个端点
… 尚未检查
工具描述中没有隐藏指令
尚未检查；我们会在检查端点时读取工具列表

与“其他”中的其他服务器比较i

星标数多于此分类中其他 32568 个服务器的 0%
发布到注册中心的时间比此分类中其他 32568 个服务器的 17% 更新
基础检查结果等于或优于此分类中其他 32568 个服务器的 31%

与注册中心里这个分类中所有活跃的服务器比较。只有当发布者拥有该仓库时，星标数才计入。

常见问题

Reply.io 安全吗？

它在 1 项适用的基础检查中通过了 0 项。我们没有阅读代码。在添加密钥或令牌之前，请先看看这个服务器要求哪些权限，并使用只具备所需权限的密钥。

它需要哪些密钥或代码？

你需要填写：X-API-Key

它在我的电脑上运行还是在云端？

它在作者的云端运行。通过互联网连接，无需安装，但你发送的内容会经过作者的服务器。

它还在维护吗？

GitHub 仓库最后更新于–，最新版本（1.0.0）于4个月前进入注册中心。

哪些应用可以使用它？

支持 MCP 的应用，例如 Claude Desktop、Claude Code、Cursor 和 VS Code。它通过 URL 连接，所以接受远程 MCP 服务器的应用也可以使用，例如开启开发者模式的 ChatGPT。

有什么可以替代它？

同一分类中有：Spec Driven Development, Remote Desktop Commander, FlyDocs

替代选择

同一分类中其他作者的服务器，按星标数排序。

如何看卡片：★ GitHub 星标i三大支柱：评价 · 检查 · 使用情况i更新 = 最新版本发布到注册中心的时间i

Spec Driven Development

作者：formulahendry

MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.

在你的电脑上运行★ 440MIT14个月前更新

安装

Remote Desktop Commander

作者：desktop-commander

Remote Desktop Commander: use files, terminal and processes on your own computers from any AI app.

在云端运行★ 211昨天更新

安装

FlyDocs

作者：flydocs.ai

Spec-driven development workflow: your agent works issues, sessions and the status lifecycle.

在你的电脑上运行今天更新

安装

作者的 README

打开 README

按原文从作者的仓库复制。PopNx MCP 没有审查或背书其内容。HTML、图片和脚本已被移除。链接会离开本站。

Reply.io MCP Server

Connect Reply.io to your AI client and let it work across your team and workspace.

reply.io · Documentation

Overview

A remote MCP server at mcp.reply.io that lets MCP-compatible AI clients — Claude, ChatGPT, Cursor, Codex, and more — work with your Reply.io CRM and AI SDR. One connection gives an agent a catalog of tools spanning contacts, sequences, inbox, tasks, analytics, and the AI SDR ("Jason"), so it can read and act on Reply data without making direct HTTP requests. The tools run the same logic as the Reply.io API and the in-app AI assistant.

No local installation required. Free for all Reply.io users.

📖 Full documentation: docs.reply.io/reply-mcp

Where this fits

Reply's agentic toolkit is three pieces. They are complementary, not alternatives:

	What it is	When you want it
[reply CLI](https://github.com/reply-team/reply-cli)	`npm i -g reply-cli` — authenticated access to every v3 endpoint via `reply api`, with Reply's own skills in the package (`reply skills install`)	Your agent has a shell. This is the complete surface.
Reply MCP (this repo)	A curated tool catalog over `mcp.reply.io`	Your client speaks MCP and has no shell — desktop apps, hosted assistants.
[Jason](https://github.com/reply-team/jason-ai)	Outbound expertise as markdown skills — what to do, in what order, with what guardrails — and a local runtime that keeps campaign work durable (`jason skills install`)	Your agent knows how to call things but not what to run.

MCP gives your agent tools. Skills give it judgement. Most setups want both; a shell-capable agent can do everything through the CLI alone.

The shortest path that works: install the CLI, reply auth login, reply skills install for the Reply skills, and add Jason's skills (jason skills install) for the judgement, then talk to your agent in plain words. MCP is optional — add it when your client has no shell.

Connect

Reply MCP is a remote server at https://mcp.reply.io. Authenticate with a scoped personal API key (recommended) or OAuth, then add it to your client.

Claude Code

claude mcp add --transport http reply-mcp https://mcp.reply.io \

--header "Authorization: Bearer usk_...your_key..."

Desktop apps (Claude, ChatGPT)

Add Reply MCP as a remote / custom connector using the URL https://mcp.reply.io and authorize with OAuth.

Create and scope a key under Settings → API Key in your Reply.io dashboard.

Note: Passing credentials in the URL (for example ?api_key=...) is not supported. Always authenticate with the Authorization header or OAuth.

Full connection guide (Codex, mcp-remote proxy, OAuth): docs.reply.io/reply-mcp#connect
Authentication & keys: docs.reply.io/reply-mcp#authentication

Authentication & scopes

Reply MCP authenticates as a single Reply.io user, via one of two methods:

Personal API key (recommended) — passed as Authorization: Bearer . A personal key is scoped: you grant it only the permissions your agent needs, and it can call only the tools those permissions cover.
OAuth — authorize directly from the client, no key to copy. OAuth grants the full tool catalog (all scopes).

Each tool requires a single scope such as contacts:read or sequences:operate; a key can call a tool only if it holds that scope. Act-on-behalf keys (Team and Organization keys) are not supported on this single-user connection.

See the full scope list at docs.reply.io/api-reference/authentication#scopes.

Tools

71 tools across your CRM and AI SDR. ⚠️ marks destructive tools (create / update / delete / send / assign) — these require the matching :write or :operate scope and carry a destructiveHint so conforming clients can prompt for confirmation. Call arguments (the JSON input schema) are advertised live via tools/list; the tables below describe what each tool is *for*.

Contacts & lists

Tool	Purpose
`reply_search_contacts`	Look up contacts by exact email or LinkedIn URL
`reply_filter_contacts`	List contacts by list/sequence membership or a name/email/company search
`reply_get_contact_activity`	Get a contact's activity history (emails, LinkedIn, calls, events)
⚠️ `reply_create_contact`	Create a contact
⚠️ `reply_update_contact`	Patch fields on an existing contact
⚠️ `reply_change_contact_owner`	Reassign contacts to another team member
⚠️ `reply_mark_contacts_as_replied`	Flag/unflag contacts as replied (stops automated outreach)
⚠️ `reply_blacklist_contact`	Blacklist a contact's email or whole domain
`reply_search_lists`	Resolve contact lists to their numeric IDs by name

Sequences

Tool	Purpose
`reply_search_sequences`	Find sequences by name, status, or archive state
`reply_list_schedules`	List sending schedules (business-hours windows)
`reply_get_sequence_steps`	Get a sequence's ordered steps (type, delay, variant count)
`reply_get_sequence_step_variants`	Get a step's A/B content variants
⚠️ `reply_change_status_in_sequence`	Set a contact's status within a sequence
⚠️ `reply_add_contact_to_sequence`	Enroll contacts into a sequence
⚠️ `reply_start_sequence`	Start or resume a sequence
⚠️ `reply_pause_sequence`	Pause a sequence
⚠️ `reply_assign_email_account_to_sequence`	Attach an email account (mailbox) to a sequence
⚠️ `reply_assign_linkedin_account_to_sequence`	Attach a LinkedIn account to a sequence
⚠️ `reply_assign_schedule_to_sequence`	Set a sequence's sending schedule

Analytics

Tool	Purpose
`reply_get_sequence_stats`	Email + LinkedIn performance for one sequence, over any date range
`reply_compare_sequence_performance`	Compare performance across multiple sequences side by side

Inbox

Tool	Purpose
`reply_get_inbox_emails`	List inbox threads (email + LinkedIn) with short previews
⚠️ `reply_send_inbox_reply`	Send a reply on an existing inbox thread
⚠️ `reply_change_inbox_category`	Set or clear a thread's category

Tasks

Tool	Purpose
`reply_list_my_tasks`	List your tasks, filtered by type, status, due window, or contact
⚠️ `reply_create_task`	Create a manual task
⚠️ `reply_complete_task`	Mark a task as completed

Channels & accounts

Tool	Purpose
`reply_list_email_accounts`	List workspace email accounts (status, daily limit, tags)
`reply_list_linkedin_accounts`	List connected LinkedIn accounts (status, health, tier)

Team

Tool	Purpose
`reply_search_team_members`	Resolve teammates to their numeric user IDs by name or email

Help & workspace

Tool	Purpose
`reply_diagnose`	Explain why a sequence isn't sending and how to fix it
`reply_get_app_map`	Catalog of Reply.io app areas and what the assistant can do in each
`reply_search_knowledge_base`	Search the Reply.io Help Center for how-to answers
`reply_get_knowledge_base_article`	Read a full Help Center article by slug
`reply_report_unsupported_request`	Report a missing capability to the product team

AI SDR — Offers

Tool	Purpose
`reply_list_offers`	List AI SDR offers (what Jason can pitch)
`reply_get_offer`	Get one offer in full
⚠️ `reply_create_offer`	Create a new offer
⚠️ `reply_update_offer`	Patch an existing offer
⚠️ `reply_attach_offer_to_sequence`	Attach an offer to an AI SDR sequence
`reply_generate_offer_from_website`	Draft offer content from a company website (review only, saves nothing)

AI SDR — Playbooks

Tool	Purpose
`reply_list_playbooks`	List playbooks (named sets of selling instructions)
`reply_get_playbook`	Get one playbook in full
⚠️ `reply_create_playbook`	Create a new playbook
⚠️ `reply_update_playbook`	Patch an existing playbook
⚠️ `reply_duplicate_playbook`	Duplicate a playbook
⚠️ `reply_attach_playbook_to_sequence`	Attach a playbook to an AI SDR sequence

AI SDR — Knowledge bases

Tool	Purpose
`reply_list_knowledge_bases`	List AI SDR knowledge bases
`reply_get_knowledge_base`	Get one knowledge base in full (with URL sources)
⚠️ `reply_create_knowledge_base`	Create a knowledge base
⚠️ `reply_update_knowledge_base`	Rename or set a knowledge base's instructions
⚠️ `reply_add_knowledge_base_source`	Add a web-page URL as a source
⚠️ `reply_delete_knowledge_base_source`	Remove a URL source
⚠️ `reply_attach_knowledge_base_to_sequence`	Attach a knowledge base to an AI SDR sequence

AI SDR — Reply handlers

Tool	Purpose
`reply_list_reply_handlers`	List reply handlers in a knowledge base
`reply_get_reply_handler`	Get one reply handler in full
⚠️ `reply_create_reply_handler`	Add a reply handler (how Jason answers a kind of question)
⚠️ `reply_update_reply_handler`	Patch a reply handler
⚠️ `reply_delete_reply_handler`	Remove a reply handler

AI SDR — Re-engagement cards

Tool	Purpose
`reply_list_reengagement_cards`	List re-engagement cards in a knowledge base
`reply_get_reengagement_card`	Get one re-engagement card in full
⚠️ `reply_create_reengagement_card`	Add a re-engagement card (how Jason wins back a quiet prospect)
⚠️ `reply_update_reengagement_card`	Patch a re-engagement card
⚠️ `reply_delete_reengagement_card`	Remove a re-engagement card

AI SDR — Approvals & reply mode

Tool	Purpose
`reply_list_pending_approvals`	List AI SDR draft messages waiting for approval
⚠️ `reply_approve_message`	Approve a draft and send it immediately
⚠️ `reply_bulk_approve_messages`	Approve and send multiple drafts at once (atomic)
⚠️ `reply_regenerate_message`	Ask the AI SDR to rewrite a pending draft
⚠️ `reply_reject_message`	Reject a draft and remove the contact from that sequence
⚠️ `reply_set_sequence_reply_mode`	Switch a sequence between Review and Autonomous reply mode

Example prompts

Once connected, you can ask your AI assistant things like:

"List my active sequences and how each is performing."
"Add john@acme.com to the Demo Follow-up sequence."
"Pause the Cold Outreach sequence — we're updating the copy."
"Why isn't my Q2 Launch sequence sending?"
"Show me the AI SDR drafts waiting for approval and approve the good ones."
"Compare step-by-step performance for my top two sequences."
"Create an offer for our new onboarding service and attach it to the Q3 sequence."

Learn more

Reply MCP documentation — capabilities, authentication, and connection guide.
Connecting Reply MCP to your AI model — Help Center walkthrough.
Reply MCP overview
reply CLI — the same Reply.io surface from a terminal, including every v3 endpoint via reply api.
[Jason](https://github.com/reply-team/jason

只显示了开头部分；其余内容请在仓库中阅读