plsreadme
MCP server for sharing markdown/text as beautiful, permanent web pages via plsreadme, with tools to share files/text, update and delete docs, and list previously shared links.
README
<p align="center"> <img src="https://plsreadme.com/icon.png" alt="plsreadme" width="80" /> </p>
<h1 align="center">plsreadme</h1>
<p align="center"> <strong>Paste markdown. Get a beautiful, shareable link. Done.</strong> </p>
<p align="center"> <a href="https://plsreadme.com">Website</a> · <a href="https://www.npmjs.com/package/plsreadme-mcp">MCP Package</a> · <a href="https://github.com/FacundoLucci/plsreadme/issues/new?labels=feature-request">Request a Feature</a> </p>
<p align="center"> <img src="https://img.shields.io/npm/v/plsreadme-mcp?style=flat-square&color=111827" alt="npm version" /> <img src="https://img.shields.io/badge/Cloudflare_Workers-deployed-F38020?style=flat-square&logo=cloudflare&logoColor=white" alt="Cloudflare Workers" /> <img src="https://img.shields.io/badge/MCP-compatible-10b981?style=flat-square" alt="MCP compatible" /> <img src="https://img.shields.io/github/license/FacundoLucci/plsreadme?style=flat-square" alt="License" /> </p>
The Problem
You wrote a README, a PRD, meeting notes, or an API doc in markdown. Now you need to share it with someone who doesn't have a markdown renderer, doesn't use GitHub, or just needs a clean link they can open in a browser.
plsreadme turns any markdown into a permanent, beautifully rendered web page in one step. No accounts. No sign-ups. No friction.
✨ Features
- Instant sharing — Paste markdown or upload a file, get a
plsrd.melink - Beautiful rendering — Clean typography, dark mode, mobile-responsive
- Inline comments — Readers can click any paragraph and leave feedback
- AI auto-formatting — Throw raw text at it; it comes out as clean markdown
- MCP server — Share docs directly from Claude, Cursor, VS Code, or any MCP client
- OpenClaw skill — Available on ClawHub for AI agent workflows
- Short links — Every doc gets a compact
plsrd.me/v/xxxURL - Raw access — Download the original
.mdfile from any shared link - Zero config — No API keys needed for basic usage
🚀 Quick Start
Web
Go to plsreadme.com, paste your markdown, click share.
API
curl -X POST https://plsreadme.com/api/render \
-H "Content-Type: application/json" \
-d '{"markdown": "# Hello World\n\nThis is my doc."}'
{
"id": "abc123def456",
"url": "https://plsreadme.com/v/abc123def456",
"raw_url": "https://plsreadme.com/v/abc123def456/raw",
"admin_token": "sk_..."
}
Save the admin_token — you'll need it to edit or delete:
# Update
curl -X PUT https://plsreadme.com/v/abc123def456 \
-H "Authorization: Bearer sk_..." \
-H "Content-Type: application/json" \
-d '{"markdown": "# Updated content"}'
# Delete
curl -X DELETE https://plsreadme.com/v/abc123def456 \
-H "Authorization: Bearer sk_..."
MCP (AI Editors)
Connect your editor to plsreadme and share docs with natural language:
"Share this README as a plsreadme link" "Turn my PRD into a shareable page" "Make these meeting notes into a readable link"
🔌 MCP Setup
Claude Code
claude mcp add plsreadme -- npx -y plsreadme-mcp
Cursor
Add to ~/.cursor/mcp.json:
{
"mcpServers": {
"plsreadme": {
"command": "npx",
"args": ["-y", "plsreadme-mcp"]
}
}
}
VS Code
Add to your settings.json:
{
"mcp": {
"servers": {
"plsreadme": {
"command": "npx",
"args": ["-y", "plsreadme-mcp"]
}
}
}
}
Claude Desktop
Add to claude_desktop_config.json:
{
"mcpServers": {
"plsreadme": {
"command": "npx",
"args": ["-y", "plsreadme-mcp"]
}
}
}
Windsurf
{
"mcpServers": {
"plsreadme": {
"command": "npx",
"args": ["-y", "plsreadme-mcp"]
}
}
}
Remote MCP (zero install)
Some clients support remote MCP endpoints directly:
https://plsreadme.com/mcp
add-mcp
npx add-mcp plsreadme-mcp
OpenClaw
clawhub install plsreadme
Docker (for MCP registries / listing checks)
Build and run the stdio MCP server in a clean container:
docker build -t plsreadme-mcp:local .
docker run --rm -i plsreadme-mcp:local
The containerized server uses stdio (no ports, no env vars required).
🛠 MCP Tools
| Tool | What it does |
|---|---|
plsreadme_share_file |
Share a local file by path → returns shareable link. Re-sharing updates the same link. |
plsreadme_share_text |
Share markdown or plain text directly → returns shareable link |
plsreadme_update |
Update an existing doc with new content (by ID or file path) |
plsreadme_delete |
Delete a shared doc permanently (by ID or file path) |
plsreadme_list |
List all documents you've shared from this project |
Prompts:
share-document— Guided flow to share content as a readable linkrefactor-and-share— Uses your AI model to refactor raw text into polished markdown, then shares it
Plain text input? No problem — the MCP auto-structures it into markdown, or you can use the refactor-and-share prompt to leverage your AI's reasoning for a polished result.
.plsreadme Record File
The MCP server tracks your shared documents in a .plsreadme JSON file in your project root. This stores document IDs, URLs, and admin tokens needed for editing and deleting.
⚠️ Add .plsreadme to your .gitignore — it contains admin tokens. The tool will warn you if it's missing.
🏗 Architecture
Built on Cloudflare's edge stack for speed everywhere:
┌─────────────┐ ┌──────────────────┐ ┌─────────┐
│ Web / API │────▶│ Cloudflare │────▶│ R2 │
│ MCP Client │ │ Workers (Hono) │ │ (docs) │
└─────────────┘ └──────────────────┘ └─────────┘
│
┌──────┴──────┐
│ D1 │
│ (metadata) │
└─────────────┘
- Hono — Lightweight web framework on Workers
- Cloudflare D1 — SQLite at the edge for metadata, comments, analytics
- Cloudflare R2 — Object storage for markdown documents
- Durable Objects — Stateful MCP server endpoint
- Workers AI — Optional fallback for text-to-markdown conversion
📁 Project Structure
plsreadme/
├── worker/
│ ├── index.ts # Main worker entry
│ ├── routes/
│ │ ├── docs.ts # Document creation & rendering
│ │ ├── comments.ts # Inline commenting system
│ │ ├── convert.ts # AI text→markdown conversion
│ │ ├── analytics.ts # View tracking
│ │ ├── links.ts # Short link handling
│ │ └── waitlist.ts # Waitlist & notifications
│ ├── mcp-agent.ts # Remote MCP server (Durable Object)
│ └── types.ts # TypeScript types
├── packages/
│ └── mcp/ # npm package: plsreadme-mcp
│ └── src/index.ts # MCP server (stdio transport)
├── public/ # Static assets & landing pages
├── db/
│ └── schema.sql # D1 database schema
├── skill/
│ └── plsreadme/ # OpenClaw agent skill
└── wrangler.jsonc # Cloudflare Workers config
🔧 Development
# Install dependencies
npm install
# Run locally
npm run dev
# Deploy
npm run deploy
# Database migrations
npm run db:migrate
MCP package release
plsreadme-mcp is published from packages/mcp by pushing an mcp-v* tag (see .github/workflows/publish-mcp.yml).
cd packages/mcp
npm version patch # or minor/major
cd ../..
git add packages/mcp/package.json packages/mcp/package-lock.json
VERSION=$(node -p "require('./packages/mcp/package.json').version")
git commit -m "chore(mcp): release v${VERSION}"
git tag "mcp-v${VERSION}"
# push commit + tag from your machine to trigger npm publish workflow
Environment Variables
Set via wrangler secret put:
| Variable | Required | Description |
|---|---|---|
OPENAI_API_KEY |
No | OpenAI key for /api/convert text→markdown |
DISCORD_WEBHOOK_URL |
No | Waitlist signup notifications |
DISCORD_LINK_WEBHOOK_URL |
No | New link creation notifications |
RESEND_API_KEY |
No | Email notifications |
NOTIFICATION_EMAIL |
No | Email recipient for notifications |
The core sharing functionality requires zero configuration. AI conversion and notifications are optional add-ons.
📊 Limits
| Limit | Value |
|---|---|
| Max document size | 200 KB |
| Upload rate limit | 30/hour per IP |
| AI convert rate limit | 10/hour per IP |
| Link lifetime | Permanent |
🤝 Contributing
Feature ideas? Bug reports? Open an issue.
PRs welcome for bug fixes and improvements.
📄 License
MIT — do whatever you want with it.
<p align="center"> <sub>Built by <a href="https://github.com/FacundoLucci">Facundo Lucci</a></sub> </p>
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。