Supabase
MCP server for interacting with the Supabase platform
作者提供的说明(英文原文)
Neon Postgres projects, branches, SQL, and migrations from your MCP client.
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。
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": {
"pgcloud": {
"command": "npx",
"args": [
"-y",
"@reneturcios/pgcloud",
"mcp"
]
}
}
}在终端中运行这条命令
claude mcp add pgcloud -- npx -y @reneturcios/pgcloud mcp粘贴到项目中的 .cursor/mcp.json,或 ~/.cursor/mcp.json
{
"mcpServers": {
"pgcloud": {
"command": "npx",
"args": [
"-y",
"@reneturcios/pgcloud",
"mcp"
]
}
}
}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": {
"pgcloud": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@reneturcios/pgcloud",
"mcp"
]
}
}
}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6日
注册名称已通过 GitHub 账户 CloudCorpRecords 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。
注册中心没有列出必需的密钥,不过有些服务器会在首次使用时要求你登录。
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(0.1.0)于5天前进入注册中心。
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, Docker。
同一分类中有:Supabase, MongoDB MCP Server, Elasticsearch MCP Server
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 for interacting with the Supabase platform
MongoDB Model Context Protocol Server
MCP server for interacting with Elasticsearch
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。
pgcloud puts a Neon Postgres account within reach of an MCP client or a terminal. Projects and their branches can be listed, described, created, and deleted; compute endpoints and organizations can be inspected; and any database in a project can be queried directly — single statements, multi-statement transactions, table and column introspection, slow-query reports, and execution plans. Schema changes follow Neon's two-phase pattern: a migration or tuning session is prepared on a temporary branch, exercised there, and only then completed onto the original branch.
A Neon API key is required. Put it in the NEON_API_KEY environment variable; it is sent to the Neon API as a Bearer token and never appears in output. When the variable is missing, every call fails with an error that names it.
npx @reneturcios/pgcloud list_projects
Or run the container image:
docker run -i -e NEON_API_KEY ghcr.io/cloudcorprecords/pgcloud
The container defaults to the MCP server (see below). Pass CLI arguments to use it as a CLI instead.
Every MCP tool is also a CLI subcommand. Tool arguments are passed as one JSON object with --json:
export NEON_API_KEY=<your Neon API key>
pgcloud list_projects pgcloud list_projects --json '{"limit": 5, "search": "billing"}' pgcloud describe_project --json '{"projectId": "snowy-pond-12345678"}' pgcloud create_branch --json '{"projectId": "snowy-pond-12345678", "branchName": "dev"}' pgcloud run_sql --json '{"projectId": "snowy-pond-12345678", "sql": "select now()"}' pgcloud get_connection_string --json '{"projectId": "snowy-pond-12345678", "pooled": true}'
pgcloud --version
Results print as JSON. Exit codes: 0 on success, 1 when an API or SQL call fails (including a missing NEON_API_KEY), 2 for usage errors.
One bin serves both faces. The mcp subcommand starts a stdio MCP server:
npx -y @reneturcios/pgcloud mcp
MCP client configuration:
{
"mcpServers": { "pgcloud": { "command": "npx", "args": ["-y", "@reneturcios/pgcloud", "mcp"], "env": { "NEON_API_KEY": "your Neon API key" } } } }
| Tool | Input (required in bold) | What it does |
|---|---|---|
list_projects | limit, search, org_id | Lists projects in the account. |
describe_project | `projectId` | Describes one project. |
create_project | name, region_id, org_id | Creates a project. |
delete_project | `projectId` | Deletes a project and its resources. |
create_branch | `projectId`, branchName, parentBranchId | Creates a branch with a read-write compute. |
delete_branch | `projectId`, `branchId` | Deletes a branch. |
describe_branch | `projectId`, `branchId` | Branch record plus its databases. |
list_branch_computes | `projectId`, branchId | Lists compute endpoints. |
list_organizations | search | Lists organizations the user can access. |
get_connection_string | `projectId`, branchId, databaseName, roleName, pooled | Resolves a Postgres connection string. |
run_sql | `sql`, `projectId`, branchId, databaseName, roleName | Runs one SQL statement. |
run_sql_transaction | `sql` (array), `projectId`, … | Runs statements in one transaction. |
get_database_tables | `projectId`, branchId, databaseName | Lists tables via information_schema. |
describe_table_schema | `projectId`, `tableName`, … | Lists a table's columns. |
list_slow_queries | `projectId`, limit, … | Slowest queries from pg_stat_statements. |
explain_sql_statement | `sql`, `projectId`, … | Returns the EXPLAIN plan for a statement. |
prepare_database_migration | `migrationSql`, `projectId`, … | Runs migration SQL on a temporary branch. |
complete_database_migration | `migrationId`, `projectId`, … | Applies a prepared migration to the original branch. |
prepare_query_tuning | `sql`, `projectId`, … | Opens a tuning session on a temporary branch with the plan. |
complete_query_tuning | `tuningId`, `projectId`, `applyChanges` | Closes a tuning session, applying or discarding. |
Management calls go to https://api.neon.tech/api/v2. SQL calls first resolve the branch's connection URI from the API, then POST the statement to the database host's HTTP SQL endpoint with the URI in the Neon-Connection-String header; SQL text is passed through as given. Migration and tuning sessions are held in the server process, so a session must be completed by the same process that prepared it.
npm install
npm run build # tsc -> dist/ npm test # node test runner over test/*.test.ts, no network (fake fetch fixtures) npm run dev # run the CLI from source via tsx
MIT — see LICENSE.