Linkly MCP Server
Linkly MCP 服务器
Linkly(URL 缩短与链接管理平台)的官方模型上下文协议(MCP)服务器。
本仓库为托管在 https://mcp.linklyhq.com 的 MCP 服务器提供支持。 这是通过 MCP 将 AI 助手连接到 Linkly 的唯一受支持方式。设置说明请参阅 https://linklyhq.com/support/mcp-server。
⚠️
linkly-mcp-servernpm 包已弃用。 自托管 / API 密钥流程不再维护。所有用户都应迁移到托管服务器——它使用带 PKCE 的 OAuth 2.1,无需 API 密钥或本地安装,并且始终运行最新的工具集。npm 包将继续存在于注册表中以保持向后兼容,但不会获得新工具或错误修复,并且将来可能会被取消发布。
文档
有关完整文档、设置指南和示例,请访问:https://linklyhq.com/support/mcp-server
Related MCP server: filoo MCP server
功能特性
创建、更新和删除短链接
查看点击分析和统计
管理自定义域名和网站图标
配置点击事件 Webhook
搜索和筛选链接
导出点击数据
使用托管服务器
服务器 URL: https://mcp.linklyhq.com
将您的 MCP 客户端连接到托管服务器,并在浏览器打开时登录 Linkly。就这样——无需 API 密钥,无需安装,无需配置漂移。
Claude Desktop
{
"mcpServers": {
"linkly": {
"command": "npx",
"args": ["-y", "mcp-remote", "https://mcp.linklyhq.com"]
}
}
}Claude Code(CLI)
claude mcp add --transport http linkly https://mcp.linklyhq.comChatGPT Desktop
{
"servers": {
"linkly": {
"type": "url",
"url": "https://mcp.linklyhq.com"
}
}
}可用工具
工具 | 描述 |
| 创建新的短链接 |
| 更新现有链接 |
| 删除链接 |
| 获取链接详情 |
| 列出工作区中的所有链接 |
| 按名称、URL 或备注搜索链接 |
| 获取时间序列点击数据 |
| 按维度分组获取点击数据 |
| 获取最近的点击数据 |
| 导出详细的点击记录 |
| 列出自定义域名 |
| 添加自定义域名 |
| 移除自定义域名 |
| 更新域名网站图标 |
| 列出工作区 Webhook |
| 添加 Webhook |
| 移除 Webhook |
| 列出链接专属 Webhook |
| 添加链接 Webhook |
| 移除链接 Webhook |
链接
支持
如需帮助和支持,请访问 Linkly 支持 或发送邮件至 support@linklyhq.com。
发布到 MCP 注册表(维护者)
仓库根目录下的 server.json 是 com.linklyhq/linkly 服务器的官方 MCP 注册表清单。它是仅远程的:它将客户端指向托管服务器,并且有意不附带任何包(已弃用的 npm 发行版已在 v2.0.0 中移除)。
发布需要对 linklyhq.com 具有 DNS 访问权限——com.linklyhq 命名空间通过域名顶点上的 DNS TXT 记录进行验证:
# 1. Generate an Ed25519 key + TXT record (macOS: use openssl@3; system LibreSSL lacks Ed25519)
openssl genpkey -algorithm Ed25519 -out key.pem
PUBLIC_KEY="$(openssl pkey -in key.pem -pubout -outform DER | tail -c 32 | base64)"
echo "linklyhq.com. IN TXT \"v=MCPv1; k=ed25519; p=${PUBLIC_KEY}\"" # add at the APEX of linklyhq.com
# 2. Authenticate the namespace + publish
PRIVATE_KEY="$(openssl pkey -in key.pem -noout -text | grep -A3 'priv:' | tail -n +2 | tr -d ' :\n')"
mcp-publisher login dns --domain linklyhq.com --private-key "$PRIVATE_KEY"
mcp-publisher publish有关详细信息,请参阅注册表身份验证指南。
更新 ChatGPT 应用(维护者)
chatgpt-app-submission.json 是 OpenAI Apps SDK 提交导入——它为全部 25 个工具预填了逐工具注释和三项安全理由说明(只读 / 开放世界 / 破坏性),因此无需在仪表板中手动输入。
OpenAI 在发布时会冻结元数据快照,并且不会自动检测 MCP 更改,因此在部署服务器更改后,您需要发布新版本:
部署 worker(推送到关联分支 /
wrangler deploy)。平台仪表板 → Linkly 应用 → 创建新草稿版本(保持相同的 MCP URL
https://mcp.linklyhq.com)。在 MCP 部分,单击 扫描工具 以重新读取实时工具列表和注释。
将
chatgpt-app-submission.json拖到表单上(页面顶部)以填写理由说明。检查测试,然后提交。已发布的版本将继续针对实时服务器运行,直到新版本获得批准。
每当添加/移除工具或它们的 readOnlyHint/destructiveHint 发生变化时,请保持此文件与 worker 同步。破坏性标志仅对 5 个真正具有破坏性的工具(delete_link、delete_domain、batchDeleteLinks、unsubscribe_webhook、unsubscribe_link_webhook)为 true;创建/更新/订阅是增量的(false)。
许可证
MIT
Available Tools
20 toolscreate_domainCInspect
Add a custom domain to the workspace. The domain must be configured to point to Linkly's servers.
Inputs: name (required).
create_linkCInspect
Create a new Linkly short link. Returns the created link with its short URL.
Inputs: url (required), name, note, slug, +19 more.
delete_domainCInspect
Remove a custom domain from the workspace.
Inputs: domain_id (required).
delete_linkCInspect
Delete a Linkly link by its ID
Inputs: link_id (required).
export_clicksCInspect
Export detailed click records with full information (timestamp, browser, country, URL, platform, referer, bot, ISP, params).
Inputs: end, bots, start, country, +2 more.
get_analyticsAInspect
Get time-series click analytics data for charting. Returns click counts over time.
Inputs: end, bots, start, unique, +5 more.
get_analytics_byBInspect
Get click counts grouped by a dimension (country, platform, browser, etc.). Useful for breakdowns and top-N reports.
Inputs: counter (required), end, bots, start, +4 more.
get_clicksCInspect
Get recent click data for the workspace
Inputs: link_id.
list_domainsBInspect
List all custom domains in the workspace.
Inputs: no named parameters.
list_linksBInspect
List all links in the workspace. Returns links with click statistics.
Inputs: no named parameters.
list_link_webhooksBInspect
List all webhook URLs subscribed to a specific link.
Inputs: link_id (required).
list_webhooksAInspect
List all webhook URLs subscribed to the workspace. These receive click events for all links.
Inputs: no named parameters.
search_linksBInspect
Search for links by name, URL, or note. Returns matching links with click statistics.
Inputs: query (required).
subscribe_link_webhookCInspect
Subscribe a webhook URL to receive click events for a specific link.
Inputs: url (required), link_id (required).
subscribe_webhookCInspect
Subscribe a webhook URL to receive click events for all links in the workspace.
Inputs: url (required).
unsubscribe_link_webhookBInspect
Unsubscribe a webhook URL from a specific link's click events.
Inputs: url (required), link_id (required).
unsubscribe_webhookCInspect
Unsubscribe a webhook URL from workspace click events.
Inputs: url (required).
update_domain_faviconCInspect
Update the favicon URL for a custom domain.
Inputs: domain_id (required), favicon_url (required).
update_linkCInspect
Update an existing Linkly link by its ID
Inputs: link_id (required), url, name, note, +18 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 20 tools
Each tool has a distinct purpose with clear boundaries, targeting specific resources (domains, links, webhooks, analytics) and actions (create, delete, get, list, update, subscribe, unsubscribe, export). There is no overlap or ambiguity; for example, get_analytics and get_analytics_by serve different analytical functions, and webhook tools are clearly separated by scope (link-specific vs. workspace-wide).
Tool names follow a consistent verb_noun pattern throughout, with verbs like create, delete, get, list, update, subscribe, unsubscribe, export, and search applied to nouns such as domain, link, webhook, analytics, and clicks. There are no deviations in style (e.g., no camelCase or mixed conventions), making the set predictable and readable.
With 20 tools, the count is slightly high but reasonable for a link management and analytics domain, covering domains, links, webhooks, and analytics comprehensively. It avoids being excessive (e.g., not over 25) and each tool appears justified, though it could be streamlined slightly without losing functionality.
The tool set provides complete CRUD/lifecycle coverage for domains (create, list, delete, update_favicon), links (create, get, list, update, delete, search), and webhooks (subscribe, unsubscribe, list for both link-specific and workspace-wide). Analytics tools (get_analytics, get_analytics_by, export_clicks, get_clicks) offer robust data access, leaving no obvious gaps for agent workflows in this domain.
Maintenance
Related MCP Connectors
Create branded short links and QR codes, and pull click analytics by country, device and referrer.
Create, search and pause deep links; QR codes; analytics, taps, events; domains; webhooks.
Short links, QR codes, link and website analytics: create and edit links, read clicks and visitors.
Create, edit and track LickityClick short links, campaigns and click stats from your AI assistant.
Related MCP Servers
- AlicenseNot gradedqualityCmaintenanceEnables interaction with the Firelinks link shortening platform to create and manage short links, track click statistics, manage custom domains, and compare analytics periods through natural language.3MIT
- AlicenseAqualityDmaintenanceEnables AI agents to shorten URLs, manage links, and track click analytics through 8 first-class tools, designed for use with Claude Desktop, Cursor, and other MCP clients.946 npmMIT
- AlicenseAqualityBmaintenanceEnables users to shorten links, bulk create URLs, generate QR codes, manage and disable links, and retrieve click analytics from MCP clients like Claude, ChatGPT, Cursor, and n8n.9MIT
- AlicenseBqualityCmaintenanceEnables users to shorten links, bulk-create URLs, generate QR codes, and retrieve click analytics through MCP clients such as Claude, ChatGPT, Cursor, and n8n.940 npmMIT