PasteAI
An MCP server that lets AI publish markdown documents to a clean, themed web UI. One tool call creates a shareable doc.
README
<div align="center">
PasteAI
Human review of beautiful reports, generated by AI
AI publishes markdown documents via MCP. You read them in a clean, themed web UI.

</div>
Quickstart
Run /setup — Claude will install the binary and configure MCP. Then restart Claude Code to activate.
Or install manually:
# macOS
brew install pasteai/pasteai/pasteai
# Windows
scoop bucket add pasteai https://github.com/pasteai/scoop-pasteai
scoop install pasteai
# Linux (Debian/Ubuntu)
# Download the .deb from https://github.com/pasteai/pasteai/releases/latest
# Any platform — prebuilt binary
curl -sSL https://raw.githubusercontent.com/pasteai/pasteai/main/install.sh | sh
# With Go
go install github.com/pasteai/pasteai/cmd/pasteai@latest
Add pasteai to ~/.claude.json (global — available in all Claude Code projects):
{
"mcpServers": {
"pasteai": {
"command": "/full/path/to/pasteai",
"args": ["mcp"]
}
}
}
Use the absolute path from which pasteai or $(go env GOPATH)/bin/pasteai — Claude Code does not inherit your shell PATH. The /setup slash command and make setup resolve the path and merge this safely (preserving any existing servers).
Restart Claude Code, then ask: "Summarise what we discussed and publish it."
CLI Reference
| Command | Description |
|---|---|
pasteai setup |
Configure MCP in ~/.claude.json (interactive or via flags) |
pasteai setup -mode embedded |
Configure without prompts (non-interactive) |
pasteai doctor |
Diagnose common setup problems |
pasteai serve [flags] |
Start the HTTP server |
pasteai mcp |
Start the MCP server (used by Claude Code) |
pasteai version |
Print version |
Which mode?
| Situation | Mode |
|---|---|
| Claude Code on your laptop | Embedded — default, nothing to configure |
| Claude running in a container or remote VM | Local server — run pasteai serve on the host, set PASTEAI_URL |
| Sharing with a team or reading across devices | Remote — deploy with -base-url, use Tailscale for private access |
Three Modes
1. Embedded (zero config)
Add the MCP config above and restart Claude Code. When Claude first calls a tool, PasteAI starts a local server automatically. Documents stored at ~/.pasteai/documents.db and as plain markdown files at ~/.pasteai/documents/{id}.md — readable without the server running.
2. Local server
Run the server persistently with Docker Compose and point the MCP client at it:
# docker-compose.yml
services:
pasteai:
image: ghcr.io/pasteai/pasteai:latest
ports:
- "8080:8080"
volumes:
- ${HOME}/.pasteai:/data
restart: unless-stopped
mkdir -p ~/.pasteai
docker compose up -d
In ~/.claude.json:
{
"mcpServers": {
"pasteai": {
"command": "/full/path/to/pasteai",
"args": ["mcp"],
"env": { "PASTEAI_URL": "http://localhost:8080" }
}
}
}
Good when you want the server always running, independent of Claude Code. For systemd, launchd, or nohup alternatives, see docs/server-setup.md.
3. Remote
Point PASTEAI_URL at any PasteAI instance. Both PASTEAI_API_KEY and -base-url are required on the server for remote access to work correctly.
Start the server with:
pasteai serve -api-key <your-key> -base-url https://your-server.example.com
In ~/.claude.json:
{
"mcpServers": {
"pasteai": {
"command": "/full/path/to/pasteai",
"args": ["mcp"],
"env": {
"PASTEAI_URL": "https://your-server.example.com",
"PASTEAI_API_KEY": "your-key"
}
}
}
}
Tailscale
To read documents from your phone or another machine, use Tailscale. Start pasteai serve, then browse to http://<tailscale-ip>:8080 from any device on your tailnet. Set -base-url http://<tailscale-ip>:8080 so document links resolve correctly.
MCP Tools
publish_document — publish a markdown document, returns a shareable URL.
| Parameter | Required | Description |
|---|---|---|
title |
yes | Document title |
content |
yes | Markdown content |
author |
no | Author name |
visibility |
no | public (default) or unlisted (link-only) |
list_documents — returns recent public documents.
REST API
| Method | Path | Description |
|---|---|---|
POST |
/api/documents |
Create a document |
GET |
/api/documents |
List recent public documents |
GET |
/api/documents/{id} |
Get a document |
curl -X POST http://localhost:8080/api/documents \
-H 'Content-Type: application/json' \
-d '{"title": "My Report", "content": "# Hello\n\nThis is my report.", "author": "Claude"}'
Response includes a url field: http://localhost:8080/d/{id}
Themes
Switch themes using the selector in the top nav. Choice is saved in localStorage.
Light · Dark · Emerald · Arctic · Catppuccin Mocha · Catppuccin Latte · Catppuccin Frappé
Contributing & License
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。