Section-by-section content refresh in Cursor
This checklist is based on How DigitalOcean updates its massive library of content 3-4x faster (without sacrificing quality) featuring Anish Singh Walia, published on the AI Lab by ActiveCampaign.
Get the checklist
By the end of this checklist, you’ll have a Cursor workspace wired to live search data through Model Context Protocol (MCP) servers, with editorial context loaded as project files, ready to refresh stale articles section by section. It’s based on the workflow Anish Singh Walia, a technical writer at DigitalOcean, built to update 4–5 tutorials per day instead of 1–2.
Plan on 1–2 hours for the one-time setup, then 2–3 hours per article refresh.
Before you start, make sure you have:
- Cursor installed and signed in
- A DataForSEO account with API credentials (login + password or API token)
- A folder containing the stale article, the SEO brief, your editorial guidelines, and any supporting research
- Access to the article source files (markdown, HTML, or whatever format your CMS exports)
The workflow
Phase 1: Wire up Cursor and your MCP servers
After this phase, you’ll have: a Cursor workspace that can pull live keyword volume, search engine results page (SERP) data, and organic results on demand.
- Install Cursor and open a fresh workspace: Download from cursor.com and create a project folder for the article you’re refreshing.
- Open MCP settings: Navigate to Settings → Tools → MCP inside Cursor.
- Add the DataForSEO MCP server: Paste in your DataForSEO API token and account credentials. Anish’s note: “You just need to copy and paste a bunch of API token stuff… very straightforward.”
- Add any infrastructure MCP servers you need: Anish keeps it lean with DigitalOcean MCP for infrastructure tasks and DataForSEO MCP for search data. Skip MCP servers you won’t use weekly.
- Verify the connection: Ask Cursor to pull search volume for any test keyword. Confirm you see successful calls to the search volume API or organic live API in the action log before moving on.
Phase 2: Load editorial context into the workspace
After this phase, you’ll have: a project folder Cursor can reference as it works, with every editorial constraint visible as a file.
- Drop the stale article into the workspace: The full current version, in the format you’ll publish back to.
- Add the SEO brief as a separate file: Updated target keywords, ranking goals, and any AI search optimization notes for this piece.
- Add brand and style guidelines: Editorial voice rules, banned phrases, tone guidance. Anish references DigitalOcean’s SEO writer checklist and AI content writing guidelines.
- Add supporting research: Previous versions of the article, first-party data, expert input, and anything else Cursor should treat as ground truth.
- Open the relevant tabs in Cursor: Cursor pulls strongest context from files you have open. Keep the brief, the article, and the style guide visible.
Phase 3: Refresh section by section
After this phase, you’ll have: an edit-ready draft aligned to current SEO data, ready for human and subject-matter-expert review.
- Tag the brief and guidelines in your prompt: Reference them by filename, the way you’d tag someone in a social post, so Cursor knows which documents to follow.
- Start with the meta title, meta description, and intro:
Starter prompt
Using @[BRIEF_FILE] and @[STYLE_GUIDE], give me an updated title, meta description, and opening paragraph for this article. Use the DataForSEO MCP to pull actual search volume and SERP data for [TARGET_KEYWORD]. Apply changes directly to the file.
- Update supporting sections one at a time: Move down the article in order. Prompt for one section per turn so you can review before continuing. Anish’s rule: “If you ask it to generate the whole tutorial, it usually generates a low-quality first draft, and most of it is AI slop. I do it step by step.”
- Generate FAQ ideas from real user questions: Pair Cursor with AlsoAsked and Answer the Public to surface questions readers actually ask, then prompt Cursor to draft FAQ entries that match those queries.
- Run a cleanup and refinement pass: Ask Cursor to check the full article for tone drift, repetition, and gaps against the brief. Reject overly broad rewrites and narrow the scope back to specific sections.
- Hand off to a subject matter expert: With the rote updates done, route the draft to the engineer, analyst, or specialist who can validate the technical claims and add expert insight.
Quick reference
- Total time: 1–2 hours one-time setup, then 2–3 hours per article (down from a full working day)
- Tools needed: Cursor, DataForSEO MCP server, AlsoAsked or Answer the Public, your existing editorial brief and style guide
- Key output: A refreshed article aligned to live search data, with rote updates handled by Cursor and the technical sections reserved for human expert review
Related
More data from the AI Lab.