Cms MCP server

@writavo/mcp-server

Your Writavo Site's content, inside an AI assistant. Sign in from the browser, bring an existing blog across, draft a post, set its category and byline, attach an image, schedule it, publish it, all from a conversation.

It speaks the Model Context Protocol over stdio, so it runs as a subprocess of your client rather than as a service you host.

Get started

Add the server to your MCP client. No key is needed to start:

{
  "mcpServers": {
    "writavo": {
      "command": "npx",
      "args": ["-y", "@writavo/mcp-server"]
    }
  }
}

Then ask the assistant to sign you in to Writavo. It calls login, which gives you a link and a short code. Open the link, sign in (or create an account and finish onboarding; choose "Bring my existing articles" if you are moving a blog), pick the Site, and approve. The assistant picks the key up by itself, with no restart and nothing to copy.

The key is made on your machine and only its hash is sent for approval. It is limited to the Site you chose, never carries key or webhook management, expires after 90 days, and is saved to ~/.config/writavo/credentials.json ($XDGCONFIGHOME is honoured; %APPDATA%\writavo on Windows) with owner-only permissions. logout deletes it; revoke the key itself at .

Or use a key you created

{
  "mcpServers": {
    "writavo": {
      "command": "npx",
      "args": ["-y", "@writavo/mcp-server"],
      "env": { "WRITAVO_API_KEY": "wv_sk_your_key_here" }
    }
  }
}

Create the key at and give it only the scopes you want the assistant to have. WRITAVOAPIKEY always takes precedence over a browser sign-in. A key never outranks the person who made it: narrowing your own permissions narrows every key you created, on the next request.

Without a key the server still starts, and login, getapidocs and the import format description all work, so you can read the whole API reference before signing up.

What it can do

Thirty six tools are compiled from Writavo's published OpenAPI specification, plus seven written by hand:

presigned upload in one call so the assistant does not have to orchestrate it.

• Articles. List, read, create, update, delete, publish, unpublish, schedule, cancel a schedule.
• Taxonomy and people. Categories, tags and authors: list, read, create, update, delete.
• Media. List, read, update, delete, and upload_media, which drives the whole three step

needs no key at all.

• Pipeline. Trigger a run, read its status, read the queue.
• Meta. Site information, content types, plan usage and balances, and getapidocs, which

happens on Stripe's page in your browser, never in the chat. The CMS (storing, publishing and importing content) is pay-as-you-go and needs no plan; a plan buys the AI article pipeline.

• Sign-in. login, login_status and logout, described above.
• Plans. startplanpurchase returns the billing link with a plan preselected. Payment

• Import. import_content brings an existing blog in from a file (below).

Importing a blog

The migrate-content prompt walks an assistant through the whole move: sign in, inspect your current system with the access you already have, write an import file, dry run, fix, import, and verify. Slugs are kept so URLs do not change, published posts keep their original publication dates, drafts stay drafts, and images are copied into your media library.

The file is the Writavo Import Format v1, a single JSON document:

Shortened. The full README is on GitHub.