Spec Driven Development
MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.
作者提供的说明(英文原文)
Manage companies, plans, features, and billing through SchematicHQ
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。
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。
选择你使用的 AI 应用,然后复制设置。
SCHEMATIC_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
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": {
"schematic": {
"command": "npx",
"args": [
"-y",
"@schematichq/schematic-mcp"
],
"env": {
"SCHEMATIC_API_KEY": "your-value"
}
}
}
}在终端中运行这条命令
claude mcp add schematic --env SCHEMATIC_API_KEY=your-value -- npx -y @schematichq/schematic-mcp运行前请先填写空着的值。
粘贴到项目中的 .cursor/mcp.json,或 ~/.cursor/mcp.json
{
"mcpServers": {
"schematic": {
"command": "npx",
"args": [
"-y",
"@schematichq/schematic-mcp"
],
"env": {
"SCHEMATIC_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。
粘贴到项目中的 .vscode/mcp.json
{
"servers": {
"schematic": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@schematichq/schematic-mcp"
],
"env": {
"SCHEMATIC_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。
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。
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
注册名称已通过 GitHub 账户 SchematicHQ 验证,该账户拥有此仓库
我们还没有从 GitHub 读取这个仓库
我们还没有从 GitHub 读取这个仓库
它在你的电脑上运行,没有互联网端点
在你的电脑上运行,改用本地设置
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。
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。
你需要填写:SCHEMATIC_API_KEY
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)。
GitHub 仓库最后更新于–,最新版本(0.4.0)于4个月前进入注册中心。
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。
同一分类中有:Spec Driven Development, Remote Desktop Commander, AvaHR
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
MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.
Remote Desktop Commander: use files, terminal and processes on your own computers from any AI app.
Read hiring data and create inactive assessment and questionnaire drafts for human review.
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。
[!WARNING]
Deprecated. Schematic now hosts an MCP server at
https://api.schematichq.com/mcp— see the Schematic MCP documentation to connect. This local stdio server still works but no longer receives new tools or fixes.
An MCP server that connects AI assistants to Schematic -- the platform for managing billing, plans, features, and entitlements.
Use this server to let Claude, Cursor, or any MCP-compatible client look up companies, manage plan entitlements, set overrides, create features, and more -- all through natural language.
npm i @schematichq/schematic-mcp
claude mcp add --transport stdio --scope project schematic --env SCHEMATIC_API_KEY=your-secret-api-key-here -- npx @schematichq/schematic-mcp
Replace your-secret-api-key-here with your actual Schematic API key.
Add to your Claude config (~/Library/Application Support/Claude/claude_desktop_config.json on macOS):
{
"mcpServers": { "schematic": { "command": "npx", "args": ["-y", "@schematichq/schematic-mcp"], "env": { "SCHEMATIC_API_KEY": "your-secret-api-key-here" } } } }
Add to your Cursor MCP config (Settings > MCP Servers, or edit the file directly):
~/.cursor/mcp.json~/.cursor/mcp.json%USERPROFILE%\.cursor\mcp.json{
"mcpServers": { "schematic": { "command": "npx", "args": ["-y", "@schematichq/schematic-mcp"], "env": { "SCHEMATIC_API_KEY": "your-secret-api-key-here" } } } }
After adding the config, restart Cursor for the server to be picked up.
The server needs a Schematic secret API key (not the publishable key). It checks two sources in order:
SCHEMATIC_API_KEY~/.schematic-mcp/config.json{
"apiKey": "your-secret-api-key-here" }
You can find your secret API key in the Schematic dashboard.
By default the server targets the production Schematic API. To point it at a local or non-production API, set SCHEMATIC_API_URL:
{
"command": "npx", "args": ["@schematichq/schematic-mcp"], "env": { "SCHEMATIC_API_KEY": "your-secret-api-key-here", "SCHEMATIC_API_URL": "http://localhost:8080" } }
When unset, the SDK's production default is used.
| Tool | Description |
|---|---|
get_company | Look up a company by ID, name, Stripe customer ID, or custom key. Returns details, plan, trial status, and links. |
create_company | Create (upsert) a company identified by a key (keyName/keyValue), with an optional name and traits. Updates the company if the key already exists. |
get_company_plan | Get the plan a company is currently on. |
get_company_trial_info | Check if a company is on a trial and when it ends. |
count_companies_on_plan | Count how many companies are on a specific plan. |
link_stripe_to_schematic | Find the Schematic company for a Stripe customer ID, or vice versa. |
| Tool | Description |
|---|---|
list_company_overrides | List overrides by company or by feature. |
set_company_override | Set or update an override for a company on a specific feature. Supports boolean (on/off), numeric, and unlimited values. |
remove_company_override | Remove an override so the company falls back to plan entitlements. |
| Tool | Description |
|---|---|
list_plans | List all plans. |
create_plan | Create a new plan. |
add_entitlements_to_plan | Add feature entitlements to a plan. Auto-detects feature type and sets appropriate value types. |
| Tool | Description |
|---|---|
list_features | List all features. |
create_feature | Create a new feature flag. Supports boolean (on/off), event-based (metered), and trait-based types. Automatically creates an associated flag. |
| Tool | Description |
|---|---|
list_flags | List all flags with a targeting summary — for each flag: key, default value, linked feature, and whether it resolves to always-on, always-off, or targeted. Useful for auditing always-on flags (redundant to check in code) or unused flags (candidates for deletion). |
get_flag | Get full targeting detail for one flag by key: default value, every rule (type, value, priority, condition count), last-checked time, and the always-on / always-off / targeted determination. |
Once configured, try asking your AI assistant:
# Install dependencies
yarn install
yarn build
yarn dev
yarn test
MIT