Featured

Deploy OpenClaw in 60 seconds — 20% off logoDeploy OpenClaw in 60 seconds — 20% off

Launch OpenClaw on Hostinger in about 60 seconds and keep your agent live 24/7. Our referral link gives you 20% off, no coupon code needed.

Launch on Hostinger
Run your Hermes agent on Hostinger, fully managed logoRun your Hermes agent on Hostinger, fully managed

Launch Hermes on Hostinger in one click, fully managed, no VPS knowledge needed. Use code ZACAARON10 for 10% off.

Launch on Hostinger
Crawl and scrape any site into clean data, 10% off logoCrawl and scrape any site into clean data, 10% off

Firecrawl crawls and scrapes any site into clean markdown for your agent. Get 1,000 free credits, and new users get 10% off their first purchase.

Try Firecrawl free
6,000+ web scrapers for your AI agent, start free logo6,000+ web scrapers for your AI agent, start free

Apify gives your agent live web data: 6,000+ prebuilt scrapers and actors, MCP-ready. Sign up free with $5 in usage credits.

Try Apify free
One API to scrape, enrich, and extract the internet. logoOne API to scrape, enrich, and extract the internet.

Context.dev gives your agents a single API to scrape, enrich, and extract live web data — no proxies, no parsers, no maintenance.

Start building free
SetupClaw: done-for-you OpenClaw for founders & exec teams logoSetupClaw: done-for-you OpenClaw for founders & exec teams

White-glove OpenClaw for founders and exec teams (4–50+ employees): we install, harden, integrate your tools, and maintain it — secured from day one.

Get it set up for you
SEO data APIs for your agent, $1 free credit logoSEO data APIs for your agent, $1 free credit

DataForSEO gives your agent live access to SERP results, keyword data, backlinks, and on-page SEO data through one API. New accounts get a $1 credit, good for up to 20,000 keyword or backlink lookups.

Try DataForSEO free
Reach 48,000+ AI builders

A flat monthly placement in front of developers actively installing AI tools. No lock-in, cancel anytime.

Advertise here

Works with

Claude CodeClaude DesktopCursorVS CodeClineCodex CLIOpenClaw+ any MCP client

Install to Claude Code

This server doesn't publish a one-line install command. Follow the setup in the source repository.

Summary

Enables content management, search, workflow operations, and Volto blocks management for Plone CMS via REST API.

README.md

Plone MCP Server

Talk to your Plone website instead of clicking through it. Plone MCP lets AI assistants like Claude create and edit pages, publish content, search the site, and manage translations on your behalf, in plain language - no coding required to use it, and nothing to change on your Plone site to enable it.

It's built on the Model Context Protocol (MCP), an open standard that lets AI assistants safely connect to external tools and data. Plone MCP exposes Plone's REST API as a set of MCP tools, so any MCP-compatible client - Claude Desktop, Claude Code, and others - can drive your site, and developers can script, automate, or build on top of the same tools.

Quickstart

Requires Node.js 22+. Add this to Claude Desktop's config file, then restart Claude Desktop:

  • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
  • Windows: %APPDATA%\Claude\claude_desktop_config.json
{
  "mcpServers": {
    "plone": {
      "command": "npx",
      "args": ["-y", "@plone/mcp"]
    }
  }
}

Now ask Claude to connect to your Plone site, e.g. "Connect to https://demo.plone.org as admin/admin".

Prerequisites

  • Node.js 22+ - Required to run the server (^20.19.0 || >=22.12.0; install: brew install node on macOS or from nodejs.org)
  • Plone 6.0+ site with REST API - The CMS you'll be connecting to

pnpm is only needed if you want to clone the repo and develop locally (see Local Development). The recommended setup below uses npx and doesn't require cloning anything.

Transports

The server ships two entry points:

  • STDIO (plone-mcp bin / dist/stdio-server.js) - for local MCP clients such as Claude Desktop.
  • HTTP (dist/http-server.js) - a streamable-HTTP server with per-session state, listening on PORT (default 3001) at /mcp. Start it with make start.

Quick Start using Claude Desktop as an example

The @plone/mcp package is published on npm, so there's nothing to install or build - npx fetches and runs it on demand.

  1. Configure Claude Desktop

Add to Claude's configuration file:

  • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
  • Windows: %APPDATA%\Claude\claude_desktop_config.json

With environment variables (optional):

{
  "mcpServers": {
    "plone": {
      "command": "npx",
      "args": ["-y", "@plone/mcp"],
      "env": {
        "PLONE_BASE_URL": "https://demo.plone.org",
        "PLONE_USERNAME": "admin",
        "PLONE_PASSWORD": "admin"
      }
    }
  }
}

Without environment variables (useful if you connect to different Plone sites and prefer to pass credentials per session via plone_configure):

{
  "mcpServers": {
    "plone": {
      "command": "npx",
      "args": ["-y", "@plone/mcp"]
    }
  }
}
  1. Restart Claude Desktop
  1. Connect to Plone

Call plone_configure once per session:

// Using environment variables
plone_configure({});

// OR providing credentials/token directly to the LLM
plone_configure({
  baseUrl: "https://demo.plone.org",
  username: "admin",
  password: "admin",
});

plone_configure({
  baseUrl: "https://demo.plone.org",
  token: "eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9...",
});

Note: Arguments take precedence over environment variables.

Local Development

Clone the repo if you want to modify the server, debug it, or run it with the MCP Inspector:

git clone https://github.com/plone/plone-mcp.git
cd plone-mcp
make install
make build

Point Claude Desktop at your local build instead of the npx command:

{
  "mcpServers": {
    "plone": {
      "command": "node",
      "args": ["/absolute/path/to/plone-mcp/dist/index.js"]
    }
  }
}

Development commands:

# Install the dependencies
make install

# Build for production (compiles TypeScript and copies blocks.json)
make build

# Run the HTTP server / the STDIO server from the build
make start
make stdio

# Debug with the MCP Inspector
make inspector

# Tests (Vitest)
make test-all       # everything
make test           # unit tests only
make test-coverage  # with coverage

# Static checks
make lint           # ESLint over src/ and __tests__/
make format         # ESLint with --fix
make type-check     # tsc over sources and tests

Run make help to list every available target.

Core Features

  • Content Management: CRUD operations on all Plone content types
  • Block System: Create and manage Volto blocks
  • Search: Full-text search with filtering and sorting
  • Workflow: Manage publication states and transitions
  • Site Info: Access content types, vocabularies, and site configuration

Essential Tools

| Tool | Description | Example | | --------------------------- | ---------------------------------------- | -------------------------------------------------------------------------------------- | | plone_configure | Connect to Plone (call once per session) | plone_configure({baseUrl, username, password}) or plone_configure({}) for env vars | | plone_get_content | Get content by path | plone_get_content({path: "/news"}) | | plone_create_content | Create new content | plone_create_content({parentPath: "/", type: "Document", title: "Page"}) | | plone_update_content | Update existing content | plone_update_content({path: "/page", title: "New Title"}) | | plone_delete_content | Delete content | plone_delete_content({path: "/old-page"}) | | plone_search | Search content | plone_search({query: "news", portal_type: ["Document"]}) | | plone_transition_workflow | Change workflow state | plone_transition_workflow({path: "/page", transition: "publish"}) | | plone_get_navigation_tree | Get hierarchical site structure | plone_get_navigation_tree({root_path: "/", depth: 2}) | | plone_get_translation | List translations of a content item | plone_get_translation({path: "/en/my-page"}) | | plone_link_translation | Link existing content as a translation | plone_link_translation({path: "/en/my-page", id: "/de/meine-seite"}) | | plone_unlink_translation | Remove a translation link | plone_unlink_translation({path: "/en/my-page", language: "de"}) |

Block Management

Creating Content with Blocks

// 1. Prepare blocks (60-second TTL - meant to be used inmediatly before content creation/editing)
plone_create_blocks_layout({
  blocks: [
    {
      type: "text",
      data: { text: "Welcome to our site!" },
    },
    {
      type: "teaser",
      data: {
        href: "/about",
        title: "Learn More",
        description: "Discover what we do",
      },
    },
  ],
});

// 2. Create content (within 60 seconds), the previously prepared blocks will automatically be included in the request
plone_create_content({
  parentPath: "/",
  type: "Document",
  title: "Homepage",
});

Managing Individual Blocks

// Add a single block
plone_add_single_block({
  path: "/homepage",
  blockType: "text",
  blockData: { text: "New paragraph" },
  position: 1,
});

// Update a block
plone_update_single_block({
  path: "/homepage",
  blockId: "51176ead-7b59-402d-9412-baed46821b36", // Get ID from plone_get_content
  blockData: { text: "Updated text" },
});

// Remove a block
plone_remove_single_block({
  path: "/homepage",
  blockId: "51176ead-7b59-402d-9412-baed46821b36",
});

Available Block Types

  • text: Rich text content
  • teaser: Link preview card with image
  • \__button: Call-to-action button
  • separator: Visual divider line

Use plone_get_block_schemas() to see all block types and their properties.

Common Workflows

Create and Publish a Page

// Configure connection
plone_configure({
  baseUrl: "https://mysite.com",
  username: "editor",
  password: "secret",
});

// Create with blocks
plone_create_blocks_layout({
  blocks: [{ type: "text", data: { text: "Article content..." } }],
});
plone_create_content({
  parentPath: "/news",
  type: "News Item",
  title: "Breaking News",
});

// Publish
plone_transition_workflow({
  path: "/news/breaking-news",
  transition: "publish",
});

Search and Filter

plone_search({
  query: "annual report",
  portal_type: ["Document", "File"],
  review_state: ["published"],
  sort_on: "modified",
  sort_order: "descending",
  b_size: 10,
});

Important Notes

⚠️ Prepared blocks expire after 60 seconds - Always call plone_create_blocks_layout immediately before creating/updating content.

⚠️ Configure once per session - Run plone_configure once at the start of each session before using other tools. Once configured, you can use all other tools without reconfiguring.

Troubleshooting

| Issue | Solution | | ------------------------------------ | ---------------------------------------------------------------------------------------------------------------------------------- | | command not found when using npx | Make sure the args in your config use plone-mcp as the binary name (not plone-mcp-server) — this was renamed in the package. | | "Plone client not configured" | Run plone_configure once at the start of your session | | "Block not found" | Use plone_get_content to get valid block IDs | | Connection errors | Verify Plone URL and credentials are correct | | Blocks not applied | Call plone_create_blocks_layout immediately before create/update (60s TTL) | | TypeScript errors during local build | Run make install to ensure all dependencies are installed |

Resources

License

MIT

See related servers & alternatives →

Related MCP servers

Browse all →

Related guides

Hand-picked reading to help you choose and use Search servers.