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)
Create, test, publish, and manage Dreamlit notification workflows from AI clients.
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 …”。
适用于网页版、桌面版和手机版 Claude:
https://mcp.dreamlit.ai/mcp在终端中运行这条命令
claude mcp add --transport http dreamlit https://mcp.dreamlit.ai/mcp粘贴到项目中的 .cursor/mcp.json,或 ~/.cursor/mcp.json
{
"mcpServers": {
"dreamlit": {
"url": "https://mcp.dreamlit.ai/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": {
"dreamlit": {
"type": "http",
"url": "https://mcp.dreamlit.ai/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。
ChatGPT 只能连接互联网上的服务器:
https://mcp.dreamlit.ai/mcp菜单名称可能因套餐和版本不同而略有差异。
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。
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。
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
注册名称已通过域名 dreamlit.ai 验证(注册中心会核实域名所有权)
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
尚未检查;我们会轮流检查各个端点
尚未检查;我们会轮流检查各个端点
尚未检查;我们会在检查端点时读取工具列表
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。
GitHub 仓库最后更新于–,最新版本(0.1.0)于6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。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。
同一分类中有:Spec Driven Development, Remote Desktop Commander, Claude Checkpoint MCP
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.
MCP server that checkpoints Claude Code session context to survive crashes and token limits
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。
Dreamlit MCP lets AI clients create, inspect, test, publish, unpublish, analyze, and style Dreamlit notification workflows.
Dreamlit is a database-driven notification workflow platform for Supabase and Postgres apps. Use this MCP server when you want an AI client to help build notification workflows from outcome-oriented prompts, review and publish them through Dreamlit, or answer bounded analytics questions about notification performance.
https://mcp.dreamlit.ai/mcpThis repository contains public setup docs and MCP registry metadata for Dreamlit's hosted remote MCP server. The production backend implementation is not open source.
Use your MCP client's remote server flow and enter:
https://mcp.dreamlit.ai/mcp
The client should open a Dreamlit authorization page. Choose the workspace you want the AI client to access and approve the requested scopes.
codex mcp add dreamlit --url https://mcp.dreamlit.ai/mcp
codex mcp login dreamlit
To request every supported v1 scope explicitly:
codex mcp login dreamlit --scopes workflows:read,workflows:write,workflows:publish,analytics:read
npx @modelcontextprotocol/inspector
Then choose Streamable HTTP and use:
https://mcp.dreamlit.ai/mcp
Dreamlit exposes product-level workflow and analytics tools, not low-level graph editing or database APIs.
| Tool | Scope | Purpose |
|---|---|---|
get_status | workflows:read | Get Dreamlit guidance, workspace context, project setup, schema hints, workflow state, and app URLs. |
list_projects | workflows:read | Find accessible Dreamlit projects in the approved workspace. |
list_workflows | workflows:read | Find workflows in a project by name, description, trigger type, and pagination. |
list_brand_kits | workflows:read | Find saved brand kit ids and style summaries with pagination before applying one to generated email drafts. |
get_workflow_and_preview_url | workflows:read | Inspect a workflow draft and get the Dreamlit preview or builder URL before editing or publishing. |
get_analytics | analytics:read | Query bounded notification analytics, recipient engagement, and workflow run data with filters and cursor pagination. |
create_or_update_workflow | workflows:write | Create a new workflow draft or update an existing draft from a natural-language prompt. |
send_workflow_test | workflows:write | Confirm and send a draft email or Slack test without publishing the workflow. |
prepare_publish | workflows:publish | Validate a workflow before publishing and return confirmation fields. |
confirm_publish | workflows:publish | Publish or schedule a workflow after explicit user confirmation. |
unpublish_workflow | workflows:publish | Disable live triggers or schedules for a workflow. |
get_status first to understand the workspace, project setup, and prompting guidance.list_projects and list_workflows if the user names a project or workflow without IDs.list_brand_kits when the user wants a specific saved brand kit applied to generated email drafts.get_workflow_and_preview_url before editing an existing workflow.get_analytics for bounded analytics questions, filtering first and paginating with returned cursors for larger result sets.create_or_update_workflow to create or update a draft.prepare_publish followed by confirm_publish.Drafts are not live after authoring. Publishing is always a separate explicit step.
Ask for the outcome, not a workflow graph. Good prompts include:
Examples:
When a new organization row is inserted, send an internal Slack alert with the organization name, owner email, and plan.
Every Monday at 9 AM America/New_York, send admins a weekly usage digest summarizing active users, failed payments, and new signups.
Tomorrow at 10 AM, announce the launch to active users who opted into product updates. Include unsubscribe.
OAuth access is scoped to the workspace selected during authorization. Personal access tokens are project scoped and can be created from Dreamlit project settings.
Available scopes:
workflows:readworkflows:writeworkflows:publishanalytics:readUse the least privilege that fits the client. Read-only workflow and brand-kit clients should request only workflows:read; analytics clients should request analytics:read.
Dreamlit MCP can access workflow metadata, saved brand kit summaries, connected project context, and bounded analytics results needed to perform the requested actions. It does not require clients to send database credentials through MCP.
Analytics access is structured and paginated. Use filters and cursors for larger result sets; CSV exports and bulk dumps are not exposed through MCP.
Do not paste secrets, raw credentials, or private tokens into prompts. Use Dreamlit's app settings for integrations and credentials.
For support, contact support@dreamlit.ai.