Docs
Cursor MCP server
Connect blogr to Cursor so its agent can read your rankings, content calendar and drafts while you build your website.
How it works
Cursor's agent can use MCP servers as tools. With blogr connected, the agent that edits your website's code also knows what your blog publishes, which pages the topical map still wants, and how each article performs in Google. Ask it to build the page a silo is missing, or to fix a draft, without switching tabs.
Cursor acts as you, with your team role on each website, and can change only what your role allows.
What you need
- A blogr account. Each website's data is available while its owner has an active plan or trial.
- Cursor on the desktop, or Cursor's web agents.
Connect Cursor
- Add blogr to ~/.cursor/mcp.json for every project, or to .cursor/mcp.json for one:
{
"mcpServers": {
"blogr": {
"url": "https://blogr.ai/mcp"
}
}
}
- Open Cursor Settings, then MCP. blogr shows up with a Needs login note.
- Select Connect. Cursor opens blogr in your browser, where you sign in and choose Allow access.
- In the agent chat, ask "Using blogr, list my websites."
Things to ask Cursor
- "Which topical map pages are planned but have no route in this app? Create them with the page template."
- "Read my three newest articles from blogr and add them to the related posts block on the home page."
- "Check which articles lost position this week and show me their meta descriptions."
- "Rename the title scheduled for Friday so it matches the query it targets."
What Cursor can do
Cursor reads without asking. Every tool that changes something is marked as a write, so Cursor asks you to confirm before it runs.
| Tool | What it does | Changes data |
|---|---|---|
| list_websites | Your websites, your role on each, and whether each has a plan, a topical map and Search Console. | No |
| get_website_overview | Article counts by status, what is scheduled next, failures, map coverage and the monthly allowance left. | No |
| list_articles | Articles by status, with dates and live links. | No |
| get_article | One article in full: the Markdown body, SEO title, meta description, categories and tags. | No |
| get_topical_map | The map by silo, then each page with its query, search volume, difficulty and current rank. | No |
| get_performance | Search Console clicks, impressions, CTR and position for 7, 28 or 90 days, with top queries and pages. | No |
| get_article_impact | How recent articles are moving in Google week over week. | No |
| list_opportunities | The SEO opportunities blogr found in your Search Console data. | No |
| get_usage | Your plan and this month's article allowance. | No |
| approve_titles | Approves titles onto your publishing schedule, one per slot, the same as the approval tray. | Yes |
| decline_titles | Declines titles that are not written yet. You can approve them again later. | Yes |
| add_article_idea | Adds your own title on the next open slot, like Add title in the app. | Yes |
| reschedule_article | Moves a scheduled article to another day and pins it there. | Yes |
| update_article | Renames a title, or edits a written draft's title, body, SEO fields, categories and tags. | Yes |
What Cursor cannot do
- Publish, delete or archive articles. Publishing stays in blogr, so nothing reaches your live site from Cursor.
- Start writing an article early. Approved titles are written on their scheduled day, within your monthly allowance.
- See or change billing, website settings, team members, or your publishing connection and its tokens.
An edit to a published article is saved in blogr. The live page changes the next time you publish that article from blogr.
Team roles
Cursor follows your role on each website. Viewers can read everything they see in the app, and Cursor cannot change anything for them. Editors and admins can approve, decline, add, reschedule and edit through Cursor. Owners can do all of it.
Limits
- 120 requests per minute per account, across every assistant you connect.
- 200 changes per account per hour.
- 25 titles added through an assistant per website per day.
Past a limit, Cursor gets a message saying so and can try again later. Nothing you do in the app counts toward these limits.
Your data
- Cursor receives only what you ask for: your websites, articles, topical map, Search Console performance and opportunities.
- blogr never sees your Cursor conversations. For each request it records which tool ran, on which website, whether it worked and how long it took, and deletes that record after 90 days.
- Cursor signs in with its own access token, which never includes your password. To disconnect, remove blogr from mcp.json, or switch it off in Cursor Settings, MCP.
Once Cursor receives data from blogr, Anysphere handles it under its privacy policy. Read blogr's own privacy policy for what we store.
Troubleshooting
- Connect does nothing: update Cursor. Current versions sign in on a local callback, and blogr also accepts the cursor:// callback older builds use.
- The agent does not use blogr: check it is switched on under Cursor Settings, MCP, and name it in your prompt.
- "blogr needs an active plan or trial": start a plan or trial on the account that owns the website.
- "Your role does not allow this": ask the website owner to give you the Editor or Admin role.
Questions
How do I add blogr to Cursor?
Add https://blogr.ai/mcp as the url of a blogr entry in mcp.json, then select Connect in Cursor Settings, MCP, and sign in to blogr.
Do I need an API token for Cursor?
No. Cursor signs in with OAuth in your browser. blogr also has a REST API with tokens, but the MCP server does not need one.
Can Cursor publish articles?
No. Cursor can approve, decline, add, reschedule and edit articles. Publishing stays in blogr.