GitHub Copilot SEO, Right in VS Code

Because the config lives in .vscode/mcp.json, the whole team gets the same SEO tools the moment they open the repo. Each person still signs in with their own SEOmatic account. It is the best fit for teams that review SEO fixes like any other pull request.

An example on the Acme Plumbing demo site. You ask GitHub Copilot:

“Why isn't /drain-cleaning-cost showing up on Google?”

GitHub Copilot calls SEOmatic:

• gsc_indexing: URL Inspection for /drain-cleaning-cost
• site_pages: crawl metadata, live page
• gsc_performance: page clicks, Sep 12 to Oct 9 (28 days)

GitHub Copilot answers:

Google crawled it but chose another page instead. URL Inspection says "Duplicate, Google chose different canonical than user", and the live page declares rel=canonical to /drain-cleaning.

The cause is in components/SeoHead.tsx: it builds the canonical from the first path segment, so every /drain-cleaning-* page points to the parent. Over Sep 12 to Oct 9 (28 days), /drain-cleaning-cost got 0 clicks of its own, while /drain-cleaning ranked at 14.6 for "drain cleaning cost".

I can change SeoHead to use the full path and add a test. You review the edit before committing.

Also try:

• “Check this branch's preview for SEO regressions”
• “Which pages lost the most clicks in 28 days?”
• “Find broken internal links and fix them”
• “Group the audit issues by component”

In the Chat view, switch to Agent mode, make sure the SEOmatic tools are selected, and try:

• “Audit our live site with SEOmatic and group the issues by the component that causes them.”
• “Which of our pages lost the most clicks in the last 28 days?”
• “Check the preview deploy of this branch for SEO regressions against production.”
• “Find broken internal links on the live site and fix them in the code.”
• “Why are these URLs not indexed by Google?”
• “We relaunch next week. Which pages earn our clicks, and do the new routes keep them?”

1. You add SEOmatic to .vscode/mcp.json and sign in when VS Code prompts you.
2. In the Chat view, you switch to Agent mode and pick the SEOmatic tools.
3. Copilot calls SEOmatic for audits, Search Console data or index status, with the exact date range behind every number.
4. It changes the code behind each issue, and you review the edits before committing.
5. You open a pull request as usual, with the SEO reasoning in the description.

Commit one file and everyone on the repo gets the same setup.

1. Create .vscode/mcp.json. Add a seomatic server of type http with the URL https://app.seomatic.ai/api/mcp, as in the snippet below.
2. Sign in. Start the server when VS Code offers to, and sign in or create a free SEOmatic account when prompted.
3. Use Agent mode. Open the Chat view in Agent mode and select the SEOmatic tools.
4. Ask your first question. Try “Audit our live site with SEOmatic and list the issues our code causes.”

Tool calls only run in Agent mode. In Ask or Edit mode, Copilot will not call SEOmatic.

Pick the workspace .vscode/mcp.json. It lives in the repo next to the code it fixes, reviews like any other file, and never shares anyone's data: each person signs in to their own SEOmatic account.

This page is about GitHub Copilot inside VS Code, in Agent mode.

• Microsoft Copilot is the chat assistant from Microsoft. For SEOmatic it is an AI answer engine: you can track whether Copilot's answers name your brand with AI visibility tracking, but it is not where you connect SEOmatic.
• Cursor is a separate AI code editor with its own Marketplace plugin: Cursor SEO.

Every coding agent connects to the same SEOmatic server, so your workspace, your free questions and your approvals carry across.

Connecting is free, and so is the SEOmatic account behind it: no card. Free covers the diagnosis with no limit: Search Console reports and insights, site audits of up to 30 pages, live page analysis, and your AI visibility results.

Live market data, like keyword volumes, backlink profiles and competitor domains, uses free questions: 5 a month, plus a one-time 10 bonus questions when you connect Google Search Console. One question covers up to 5 lookups.

Paid plans remove the question limit and add the treatment: agents stage fixes on your site (titles, descriptions, content updates, technical fixes, internal links) and run campaigns and fresh AI visibility scans. Every paid plan starts with a 14-day free trial. See pricing.

Each teammate signs in with their own account and allowance, so a shared config never shares anyone's data.

Copilot's code edits go through your review and your pull requests. Anything SEOmatic would change on a live site through a connected CMS is staged, never pushed. A staged change is a proposal: it says what will change and why, and nothing touches your site until you approve it. The assistant records your decision, and it is told never to approve anything on its own judgment. Approved fixes start within seconds, the page's previous state is stored, and any applied change can be rolled back. Changes that could take a page out of Google, like a noindex, need an extra explicit confirmation.

Other coding agents: Cursor, Claude Code, Codex and Zed. Explore technical SEO fixes, broken link fixing and page indexing. In CI or a terminal? The SEOmatic CLI runs the same tools. All assistants: AI assistants. Full reference: MCP docs.

Using another agent? See Cursor, Codex, Claude Code, or every AI agent.