VibeSharing MCP Server
Connects AI coding assistants to a team prototype hub for deploying, sharing, and gathering feedback on prototypes, with context sync across sessions.
README
VibeSharing MCP Server
The missing link between your AI coding assistant and your team.
Build prototypes with Claude Code, Cursor, or any AI tool—then instantly share them with your team for feedback. No context switching. No copy-pasting URLs. Just ship and collaborate.
The Problem
You're building fast with AI. Prototypes are flying. But then:
- You deploy to Vercel and... forget to tell anyone
- Your teammate asks "where's that dashboard thing you made?"
- Feedback lives in Slack threads that disappear
- Next AI session? All that context is gone
Your AI helps you build. But it can't help you collaborate.
Until now.
The Solution
VibeSharing's MCP server connects Claude Code directly to your team's prototype hub. Every prototype you build becomes instantly shareable, trackable, and open for feedback.
You: "Post this prototype to VibeSharing"
Claude: I found these collections in your org:
1. All Hero Use Cases
2. Compliance Hub
3. Internal Tools
Which collection should this go in? And do you want a custom URL
like erg-v3-teams.vercel.app?
No guessing. No duplicates. Just a quick confirmation and your prototype is live.
What You Can Do
Ship prototypes without leaving your terminal
"Deploy this to VibeSharing in the Hero Use Cases collection as erg-v3-teams"
Claude confirms the collection, names the deployment, and handles everything—GitHub repo, Vercel deploy, team registration—in one conversation.
Check what your team thinks
"What feedback did I get on the checkout flow prototype?"
See comments, suggestions, and resolved issues without opening a browser.
Keep context alive across sessions
"Sync my CLAUDE.md (or AGENTS.md) to VibeSharing"
Your project context persists on VibeSharing, so any team member (or AI session) can pick up where you left off.
See all your work in one place
"List my prototypes"
"Search collections for hero"
Every prototype you've registered, with links and recent activity. Fuzzy search included.
Quick Start
1. Get Your Token & Connect GitHub
Sign up at vibesharing.app, then go to Account Settings to:
- Connect your GitHub account — required for Push to Deploy (gives you automatic push access to prototype repos)
- Copy your deploy token — needed for the MCP server config below
2. Configure Claude Code
claude mcp add vibesharing -s user \
-e VIBESHARING_TOKEN=vs_PASTE_YOUR_TOKEN \
-- npx -y @vibesharingapp/mcp-server@latest
Important: Replace
vs_PASTE_YOUR_TOKENwith your actual token. Use-s userso it works in all projects. Then restart Claude Code.
Or add to your Claude Code settings manually:
{
"mcpServers": {
"vibesharing": {
"command": "npx",
"args": ["-y", "@vibesharingapp/mcp-server@latest"],
"env": {
"VIBESHARING_TOKEN": "vs_your_token_here"
}
}
}
}
Using @latest ensures you always get new features and fixes automatically.
3. Start Building
That's it. Ask Claude to register prototypes, check feedback, or sync context. It just works.
Why This Matters
For Solo Builders
Stop losing track of what you've built. Every prototype is catalogued, shareable, and ready for feedback when you need it.
For Teams
Finally, visibility into what everyone's shipping. No more "hey, can you send me that link again?" Feedback is structured, threaded, and actionable.
For Design Leaders
See the full picture of your team's prototyping velocity. Guide feedback with custom topics. Keep stakeholders in the loop without endless meetings.
Available Tools
| Tool | Description |
|---|---|
resolve_target |
Start here. Fuzzy-matches collection/project names and confirms where to deploy before proceeding. |
import_repo |
Import a GitHub repo into VibeSharing with a named Vercel deployment |
deploy_files |
Deploy multi-file projects to VibeSharing with named deployments |
deploy_prototype |
Deploy a single code file directly |
register_prototype |
Register an already-deployed prototype with name, description, and URL |
list_prototypes |
List/search all prototypes in your organization |
list_collections |
List/search all collections |
get_feedback |
Get feedback and comments for any prototype |
sync_context |
Push CLAUDE.md, AGENTS.md, or project notes to VibeSharing |
create_collection |
Create a new collection |
upload_source |
Upload source code to an existing prototype |
add_context_link |
Attach reference links (Figma, PRDs, docs) to collections or projects |
list_context_links |
List reference links on a collection or project |
remove_context_link |
Remove a reference link |
verify_token |
Check that your deploy token is valid |
Environment Variables
| Variable | Required | Description |
|---|---|---|
VIBESHARING_TOKEN |
Yes | Your deploy token from VibeSharing |
VIBESHARING_URL |
No | Custom API URL (defaults to https://vibesharing.app) |
Examples
Deploy with a named URL
You: "Deploy this to VibeSharing as erg-v3-teams in the Hero Use Cases collection"
Claude: Deployed!
Live URL: https://erg-v3-teams.vercel.app
VibeSharing: https://vibesharing.app/dashboard/projects/xyz789
Your team can now view the prototype and leave feedback.
Fuzzy search for collections
You: "Put this in the hero collection"
Claude: I found "All Hero Use Cases" — is that the right collection?
Get feedback before your next session
You: "Before I continue on the dashboard, what feedback came in?"
Claude: 3 new comments on Dashboard Redesign:
- Sarah: "Love the new nav, but can we make the search more prominent?"
- Mike: "The loading states feel snappy now"
- Alex: "Can we add keyboard shortcuts?" [Resolved]
Keep your AI context in sync
You: "Sync my CLAUDE.md to the Dashboard project on VibeSharing"
Claude: Context synced! Your team can now see your project notes at:
https://vibesharing.app/dashboard/projects/abc123
Built for the AI-Native Workflow
VibeSharing isn't just another tool to check. It's infrastructure for teams building with AI:
- Smart deploy targeting — fuzzy-matches collections and projects so you don't need exact IDs
- Named deployments — deterministic URLs like
erg-v3-teams.vercel.appinstead of random hashes - Context file sync (CLAUDE.md, AGENTS.md) keeps context alive across sessions and team members
- Guided feedback topics help stakeholders give useful input
- Email notifications when prototypes update or get feedback
- Works with any deploy target — Vercel, Netlify, Replit, Lovable, v0, or paste any URL
Learn More
- VibeSharing — Sign up free
- Documentation — Full setup guide
- GitHub — Source code
Stop building in isolation. Start shipping with your team.
claude mcp add vibesharing -s user \
-e VIBESHARING_TOKEN=vs_PASTE_YOUR_TOKEN \
-- npx -y @vibesharingapp/mcp-server@latest
Important: Replace
vs_PASTE_YOUR_TOKENwith your actual token. Use-s userso it works in all projects. Then restart Claude Code.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。