Connect Storyblok to Cursor via MCP: Stories and Schemas

💡 No install? Use cloud.anythingmcp.com directly. Sign in, click Connectors → Storyblok, paste your credentials, mint an MCP API key and you're done. No Docker, no git clone, no local server.

Storyblok + Cursor

Storyblok headless CMS through the Management API: find and read stories and folders with their full block content and version history, create, update, publish and unpublish stories, read component schemas, manage asset metadata, datasource entries, workflow stages and releases, in the EU, US, Canada, Australia or China region.

Prerequisites

See the full setup instructions baked into the connector (visible in the in-app store when you select the connector). The required environment variables for this connector are:

STORYBLOK_PERSONAL_ACCESS_TOKEN, STORYBLOK_SPACE_ID, STORYBLOK_API_HOST

Step 1 — Get credentials

Setting up takes about 3 minutes. You need a Storyblok space, its numeric id and the region it was created in.

1. In Storyblok open My account > Account settings > Personal access tokens, generate a token and paste it into STORYBLOK_PERSONAL_ACCESS_TOKEN. Storyblok takes it as is in the Authorization header, without a Bearer prefix. The token acts with your own rights in every space you can open, so a user with a restricted role gives a narrower token.
2. STORYBLOK_SPACE_ID: the numeric id in Settings > Space, also in the app URL after #/me/spaces/, for example 288868932106293.
3. STORYBLOK_API_HOST: the Management API host of the region where the space was created: mapi.storyblok.com (EU), api-us.storyblok.com (US), api-ca.storyblok.com (Canada), api-ap.storyblok.com (Australia) or app.storyblokchina.cn (China). Only the host, without https://.
4. Install the connector and run storyblok_get_space: it shows name, plan, languages and story count of the space and confirms all three values.

Drafts included: the Management API reads drafts as well as published content, so no Content Delivery API token is needed.

What it reads:

• Stories and folders with storyblok_list_stories, without their content: filter by folder (starts_with), by_slugs, search, text_search, tags, filter_query on content fields, published state, workflow stage, release or trash.
• One story with its full block tree (storyblok_get_story): each block with component and _uid, plus translations, tags, workflow stage and release info. storyblok_list_story_versions shows its history with event, author and time.
• The block schemas with storyblok_list_components and storyblok_get_component: every field with type, required flag, options and allowed blocks, and the component folders.
• Asset metadata: filename URL, title, alt text, copyright, focus point, folder, tags, size and private flag; asset folders and internal tags.
• Datasources and their entries (the key in name, the value in value, optionally per dimension such as a language).
• Workflows per content type, their stages and a story's stage history; releases with their schedule and state.

What it writes:

• storyblok_create_story creates a story or a folder, saved as a draft unless publish is true. The content root needs component set to a content type block; nested blocks sit in arrays and need a _uid. The AI learns the schema first with storyblok_list_components.
• storyblok_update_story replaces the content as a whole: the AI reads the story with storyblok_get_story, changes it and sends back the full block tree. publish true also publishes.
• storyblok_publish_story and storyblok_unpublish_story, optionally for some languages only. Storyblok runs them as GET requests, but they change the live site, so they count as writes.
• storyblok_update_asset changes alt text, title, copyright, source, folder, tags and the private or locked flag; the file itself stays the same. storyblok_create_datasource_entry and storyblok_update_datasource_entry maintain key-value lists.
• storyblok_change_workflow_stage moves a story to another stage, assigns users or roles and notifies them with a comment.
• storyblok_create_release and storyblok_update_release; do_release: true publishes every story change in the release at once. A story change goes into a release with release_id on create or update.

Switched off at install: storyblok_delete_story (a folder goes with all its stories, and they leave the delivery API at once) and storyblok_delete_asset (stories that still point to its URL show a broken file). They cannot be undone through this connector, so a workspace admin turns them on when the team needs them.

Data formats: ids are numbers, uuid fields are UUIDs. full_slug is the path of a story; published and unpublished_changes tell its state. The asset filename is its public URL. Times are ISO 8601 UTC; release_at takes YYYY-MM-DD HH:mm.

Paging: stories, versions, assets, datasources and datasource entries take page (from 1) and per_page (25 by default, at most 100, or 1,000 for datasource entries). The total arrives in the total header, shown as _headers.total. A page past the end returns an empty list.

Limits: 3 requests per second on the Starter plan, 6 on higher plans; 429 means slow down.

Common errors: 401 means a wrong token, or a region host that does not match the space. 404 means a wrong space id or an unknown item. 422 means an invalid story, and the message names the field.

The connector does not upload files, does not use the Content Delivery API and does not change component schemas.

Step 2 — Install the adapter

curl -fsSL https://raw.githubusercontent.com/HelpCode-ai/anythingmcp/main/docker-compose.quickstart.yml -o docker-compose.yml
printf 'JWT_SECRET=%s\nENCRYPTION_KEY=%s\n' "$(openssl rand -hex 32)" "$(openssl rand -hex 32)" > .env
docker compose up -d

Step 3 — Add the connector in Cursor

Your server URL: in AnythingMCP, open MCP Servers → the server this connector is on, and copy its URL (https://cloud.anythingmcp.com/mcp/…). Use it wherever this guide shows YOUR_SERVER_ID.

Cursor reads MCP servers from ~/.cursor/mcp.json. Add this entry:

{
  "mcpServers": {
    "anythingmcp": {
      "url": "https://cloud.anythingmcp.com/mcp/YOUR_SERVER_ID",
      "headers": { "X-API-Key": "YOUR_MCP_API_KEY" }
    }
  }
}

1. Create an MCP API key in AnythingMCP under MCP Servers → your server → API keys, and paste it in place of YOUR_MCP_API_KEY.
2. Save the file and restart Cursor.
3. Open Cursor → Settings → MCP and check that anythingmcp is listed and connected.
4. Start chatting: the agent can now call the Storyblok tools.

Available tools

Example prompts

• "List the stories under blog/ that have unpublished changes."
• "Show me the content of the home story and which blocks it uses."
• "Fix the typo "recieve" in the FAQ story and publish it."
• "Which images in the asset library have no alt text? Suggest alt text for the first ten."
• "Move the pricing page story to the Review stage and notify the marketing role with a comment."

FAQ

Does Cursor support custom MCP servers on the free tier? Yes, MCP is available on Hobby, Pro and Business plans.

Next steps

• Connector landing page
• Connect to Claude
• Connect to ChatGPT
• Connect to Gemini
• Connect to Meta Muse
• Connect to Microsoft Copilot