MCP FreeScout
概览
作者提供的说明(英文原文)
FreeScout ticket management with AI-powered analysis, search filters, and structured outputs.
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 1 项依据公开信息检查;尚未扫描代码
- 使用情况i非常新↓ 162/周 · ★ –
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: 在你的电脑上运行
- 传输方式: stdio
- 分类: 搜索与 AI 记忆
- 你需要填写:FREESCOUT_URL, FREESCOUT_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 1 项
- 在你的电脑上运行,拥有你的用户账户的权限(文件、网络、应用)。
- 需要你自己填写:
FREESCOUT_URL,FREESCOUT_API_KEY,FREESCOUT_DEFAULT_USER_ID
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
- npm已安装 Node.js(用于 npx 命令)
你用的是哪个应用?
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 …”。
打开 Claude Desktop → Settings → Developer → Edit Config,把下面的内容粘贴到 claude_desktop_config.json
{
"mcpServers": {
"freescout": {
"command": "npx",
"args": [
"-y",
"@verygoodplugins/mcp-freescout"
],
"env": {
"FREESCOUT_URL": "your-value",
"FREESCOUT_API_KEY": "your-value",
"FREESCOUT_DEFAULT_USER_ID": "your-value"
}
}
}
}在终端中运行这条命令
claude mcp add freescout --env FREESCOUT_URL=your-value --env FREESCOUT_API_KEY=your-value --env FREESCOUT_DEFAULT_USER_ID=your-value -- npx -y @verygoodplugins/mcp-freescout运行前请先填写空着的值。
或者自己复制设置
粘贴到项目中的 .cursor/mcp.json,或 ~/.cursor/mcp.json
{
"mcpServers": {
"freescout": {
"command": "npx",
"args": [
"-y",
"@verygoodplugins/mcp-freescout"
],
"env": {
"FREESCOUT_URL": "your-value",
"FREESCOUT_API_KEY": "your-value",
"FREESCOUT_DEFAULT_USER_ID": "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。
或者自己复制设置
粘贴到项目中的 .vscode/mcp.json
{
"servers": {
"freescout": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@verygoodplugins/mcp-freescout"
],
"env": {
"FREESCOUT_URL": "your-value",
"FREESCOUT_API_KEY": "your-value",
"FREESCOUT_DEFAULT_USER_ID": "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。
这个服务器在你的电脑上运行,而 ChatGPT 只能连接互联网上的服务器。请选择其他应用。
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
- Your FreeScout instance URL (e.g., https://support.example.com)
- FreeScout API key with read/write permissions
FREESCOUT_DEFAULT_USER_IDenvDefault user ID for ticket operations
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
它在你的电脑上运行;安装后,工具列表会显示在你的 AI 应用中。
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10日
- ✓ 通过
作者与仓库一致
注册名称已通过 GitHub 账户 verygoodplugins 验证,该账户拥有此仓库
- … 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
- – 不适用
端点有响应
它在你的电脑上运行,没有互联网端点
- – 不适用
使用 OAuth 登录
在你的电脑上运行,改用本地设置
- … 尚未检查
工具描述中没有隐藏指令
它在你的电脑上运行;我们只能读取互联网端点的工具列表
与“搜索与 AI 记忆”中的其他服务器比较i
- 星标数多于此分类中其他 5832 个服务器的 0%
- 发布到注册中心的时间比此分类中其他 5832 个服务器的 34% 更新
- 基础检查结果等于或优于此分类中其他 5832 个服务器的 100%
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
MCP FreeScout 安全吗?
它在 1 项适用的基础检查中通过了 1 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:FREESCOUT_URL, FREESCOUT_API_KEY
它在我的电脑上运行还是在云端?
它在你的电脑上运行。你的 AI 应用会在需要时启动它,数据不会经过作者的服务器(除非它自己调用在线服务)。
它还在维护吗?
GitHub 仓库最后更新于–,最新版本(3.1.0)于2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。你的电脑上还需要有 Node.js。
有什么可以替代它?
同一分类中有:Context7, Codebase Memory, Exa
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:↓ 每周下载量i★ 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
Codebase Memory
Codebase knowledge graph for AI agents — 162 languages, sub-ms queries, 99% fewer tokens.
Exa
Connect AI agents to Exa for web search, content fetching, and multi-step research.
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。
FreeScout MCP Server
An MCP (Model Context Protocol) server for FreeScout helpdesk ticket management. This server provides tools to interact with FreeScout tickets, analyze issues, and manage customer responses.
Features
- 🎫 Ticket Management: Fetch, analyze, and update FreeScout tickets
- 🔍 Intelligent Analysis: Automatically analyze tickets to determine issue type, root cause, and solutions
- 💬 Draft Responses: Generate customer replies based on ticket analysis
- 📊 Advanced Search: First-class filter parameters with relative time support ("7d", "24h")
- 🔒 Type Safety: Full Zod schema validation with structured outputs
- 🔁 Reliability: Automatic retry logic with exponential backoff for transient failures
- ⚡ Protocol-ready stdio: A fresh
buildServerfactory serves both 2025-era and 2026 MCP stdio clients
v3 runtime
- Requires Node.js 24 or newer.
- Uses
@modelcontextprotocol/server2.x with Zod 4 input schemas. - Serves both the 2025 legacy handshake and the 2026 stdio protocol from the same server factory.
- Delivers
structuredContentfor stable analysis and write-operation results without declaring output schemas. - Accepts FreeScout's successful
204 No Contentupdate responses and records the user ID that initiated a ticket update.
Installation
Prerequisites
- Node.js 24 or higher
- FreeScout instance with API access enabled
Quick Start (Recommended)
The easiest way to use this MCP server is with npx:
With Claude Desktop
Add this to your Claude Desktop settings (~/Library/Application Support/Claude/claude_desktop_config.json on macOS):
{
"mcpServers": { "freescout": { "command": "npx", "args": ["@verygoodplugins/mcp-freescout@latest"], "env": { "FREESCOUT_URL": "https://your-freescout-domain.com", "FREESCOUT_API_KEY": "your-api-key-here" } } } }
With Cursor IDE
Add this to your Cursor MCP settings:
Method 1: Via Cursor Settings UI
- Open Cursor Settings (Cmd/Ctrl + ,)
- Search for "MCP"
- Click "Edit in settings.json"
- Add the MCP server configuration
Method 2: Manual Configuration Add this to your Cursor settings.json or create ~/.cursor/mcp.json:
{
"mcp": { "servers": { "freescout": { "command": "npx", "args": ["@verygoodplugins/mcp-freescout@latest"], "env": { "FREESCOUT_URL": "https://your-freescout-domain.com", "FREESCOUT_API_KEY": "your-api-key-here" } } } } }
The server communicates only with the FreeScout instance configured in its environment.
Manual Installation (Alternative)
If you prefer to install and run the server locally:
- Clone this repository:
git clone https://github.com/verygoodplugins/mcp-freescout.git
cd mcp-freescout
- Install dependencies:
npm install
- Build the TypeScript code:
npm run build
- Configure your MCP client to use the local installation:
{
"mcpServers": { "freescout": { "command": "node", "args": ["/path/to/mcp-freescout/dist/index.js"], "env": { "FREESCOUT_URL": "https://your-freescout-domain.com", "FREESCOUT_API_KEY": "your-api-key-here" } } } }
Usage with Other MCP Clients
Run the server directly:
npm start
Or in development mode with auto-reload:
npm run dev
Available Tools
Core Ticket Operations
freescout_get_ticket
Fetch a FreeScout ticket with all its details and conversation threads.
Parameters:
ticket(required): Ticket ID, number, or FreeScout URLincludeThreads(optional): Include conversation threads (default: true)
Natural Language Examples:
- "Show me ticket #12345"
- "Get the details for FreeScout ticket 34811"
- "Fetch ticket https://support.example.com/conversation/12345"
- "What's in ticket 12345?"
- "Pull up the conversation for ticket #34811"
Example:
{
"ticket": "12345", "includeThreads": true }
Example: Fetching a FreeScout ticket with conversation threads
freescout_analyze_ticket
Analyze a ticket to determine issue type, root cause, and suggested solutions.
Parameters:
ticket(required): Ticket ID, number, or FreeScout URL
Natural Language Examples:
- "Analyze ticket #12345"
- "What kind of issue is ticket 34811?"
- "Can you analyze this ticket and tell me if it's a bug?"
- "Examine ticket #12345 and determine the root cause"
- "Is this ticket a bug or feature request?"
Returns:
- Customer information
- Issue description and classification
- Code snippets and error messages
- Reproducibility status
- Root cause analysis
- Bug vs feature request vs third-party issue determination
Example: Intelligent ticket analysis with issue classification
freescout_add_note
Add an internal note to a ticket for team communication.
Parameters:
ticket(required): Ticket ID, number, or FreeScout URLnote(required): The note contentuserId(optional): User ID for the note (defaults to env setting)
Natural Language Examples:
- "Add a note to ticket #12345 saying 'Reproduced on staging'"
- "Leave an internal note on this ticket"
- "Add a team note: 'Customer confirmed fix works'"
- "Note on ticket 34811: 'Escalating to development team'"
- "Add internal documentation to this ticket"
freescout_update_ticket
Update ticket status and/or assignment.
Parameters:
ticket(required): Ticket ID, number, or FreeScout URLstatus(optional): New status ('active', 'pending', 'closed', 'spam')assignTo(optional): User ID to assign the ticket to
Natural Language Examples:
- "Close ticket #12345"
- "Mark ticket 34811 as pending"
- "Assign this ticket to user ID 2"
- "Set ticket status to active"
- "Update ticket #12345 status to closed and assign to user 1"
freescout_create_draft_reply
Create a draft reply in FreeScout that can be edited before sending. This tool lets the LLM generate the reply content and saves it directly to FreeScout as a draft. Automatically converts Markdown formatting to HTML for proper display in FreeScout.
Parameters:
ticket(required): Ticket ID, number, or FreeScout URLreplyText(required): The draft reply content (generated by the LLM, supports Markdown formatting)userId(optional): User ID creating the draft (defaults to env setting)to(optional): List of TO recipients. Omit to preserve existing recipients; pass[]to clear.cc(optional): List of CC recipients. Omit to preserve existing recipients; pass[]to clear.bcc(optional): List of BCC recipients. Omit to preserve existing recipients; pass[]to clear.
Natural Language Examples:
- "Create a draft reply for ticket #12345"
- "Draft a customer response for this ticket"
- "Generate and save a draft reply explaining the fix"
- "Write a draft response to the customer for ticket 34811"
- "Create a draft reply thanking the customer and explaining the solution"
If recipient fields are omitted, the server preserves the current conversation recipients from FreeScout when available. If FreeScout does not expose existing to recipients on the conversation, its normal default customer recipient behavior is preserved.
Markdown Support:
- Bold text:
**text**or__text__→ text - _Italic text_:
*text*or_text_→ _text_ Code:`code`→code- Numbered lists:
1. item→ proper ordered lists - Bullet lists:
- itemor* item→ proper unordered lists - Line breaks: Double newlines create paragraphs, single newlines create line breaks
Workflow:
- Use
freescout_get_ticket_contextto get customer info and ticket details - Let the LLM craft a personalized reply using Markdown formatting
- Use
freescout_create_draft_replyto save the draft in FreeScout (Markdown automatically converted to HTML) - Review and edit the draft in FreeScout before sending
Example: Draft reply workflow with personalized customer response
freescout_get_ticket_context
Get ticket context and customer information to help craft personalized replies.
Parameters:
ticket(required): Ticket ID, number, or FreeScout URL
Natural Language Examples:
- "Get context for ticket #12345 to write a reply"
- "I need customer info and ticket details for drafting a response"
- "Gather context for this ticket so I can write a personalized reply"
- "Pull customer information and issue details for ticket 34811"
- "Get ticket context to help craft a customer response"
Returns:
- Customer name and email
- Ticket subject and status
- Issue description and analysis
- Recent customer and team messages
- Analysis results (bug vs feature vs third-party issue)
freescout_search_tickets
Search for tickets across your FreeScout instance.
Parameters:
query(required): Search querystatus(optional): Filter by status ('active', 'pending', 'closed', 'spam', 'all')mailboxId(optional): Filter by specific mailbox ID (searches all mailboxes if not specified)
Natural Language Examples:
- "Search for tickets containing 'OAuth error'"
- "Find all pending tickets with 'HighLevel' in them"
- "Search for closed tickets about 'plugin conflicts'"
- "Look for tickets from customer 'victor@example.com'"
- "Find all active tickets related to 'authentication'"
- "Search for tickets in mailbox 1 containing 'bug report'"
- "Find tickets in mailbox 2 with status pending"
Search Parameters (v2.0+):
textSearch(optional): Plain text search in ticket content/subjectassignee(optional): 'unassigned' | 'any' | user_id (number)status(optional): 'active' | 'pending' | 'closed' | 'spam' | 'all'state(optional): 'published' | 'deleted'mailboxId(optional): Filter by specific mailbox IDupdatedSince(optional): ISO date or relative time like "7d", "24h", "30m"createdSince(optional): ISO date or relative timepage(optional): Page number for pagination (min: 1)pageSize(optional): Results per page (min: 1, max: 100)
Search Tips for AI Agents:
- For unassigned tickets: Use
assignee: "unassigned"withstatus: "active" - For recent tickets: Use
updatedSince: "7d"for last 7 days - For specific user: Use
assignee: 123(user ID number) - Status "active" = open/active tickets (NOT "open" - that's invalid)
- Use freescout_get_mailboxes first if filtering by mailbox
- Combine filters:
{ textSearch: "error", assignee: "unassigned", updatedSince: "24h" }
freescout_get_mailboxes
Get a list of all available mailboxes in your FreeScout instance.
Parameters: None
Natural Language Examples:
- "Show me all available mailboxes"
- "List the mailboxes in FreeScout"
- "What mailboxes are configured?"
- "Get mailbox information"
Workflow Examples
Basic Ticket Analysis
// Analyze a ticket to understand the issue
await mcp.callTool('freescout_analyze_ticket', { ticket: '12345', });
Complete Ticket Response Workflow
// 1. Analyze the ticket to understand the issue
const analysis = await mcp.callTool('freescout_analyze_ticket', { ticket: '12345', });
// 2. Get ticket
只显示了开头部分;其余内容请在仓库中阅读