LinkedditLinkeddit

n8n + Linkeddit

Build an n8n weekly SEO report that also shows AI answer visibility

An n8n weekly SEO report workflow normally pulls seven days of Google Search Console data every Monday and emails clicks, impressions, CTR and average position. That covers Google's blue links only. Linkeddit adds the other half: where AI answer engines name a competitor instead of you. Claude uses the n8n connector to build the workflow and the Linkeddit connector to supply those rows, and you read the draft before it is published.

Updated 2026-10-29

Connect both

Your AI client sits in the middle. Linkeddit supplies the data; the other connector makes the change.

Linkeddit connector

Add https://mcp.linkeddit.com/mcp as a connector and sign in. Claude Code: claude mcp add --transport http linkeddit https://mcp.linkeddit.com/mcp.

Setup guide

n8n connector

  1. In n8n, open Settings, then Instance-level MCP, and turn on Enable MCP access. n8n documents this for version 2.18.3 or higher, on every plan including Community.
  2. Copy the Server URL from the OAuth tab, add the n8n connector in Claude and authorize it. Choose which workflows are available to MCP, and require approval before read and write tools run.
  3. Add the Linkeddit connector at https://mcp.linkeddit.com/mcp, and decide who receives the report so Claude never guesses a recipient.
n8n docs

Works in: Claude, Claude Code.

The prompt

Paste into Claude

Use the Linkeddit connector and the n8n connector. 1) With Linkeddit, call answer_runs_get for my project and list the audit runs from the last seven days with their status and completion date. 2) Call compete_brief_get. If it is empty, say so and leave it out. 3) In n8n, call search_workflows for an existing weekly SEO report and open it with get_workflow_details. 4) If none exists, call create_workflow_from_code to draft one with a Monday schedule, a Search Console step, a data table named ai_visibility_weekly and an email step. Add this week's runs to the table with add_data_table_rows, and mark any partial run as partial. 5) Call validate_workflow and test_workflow, then show me the workflow and the test result. Do not call publish_workflow until I say so.

Step by step

  1. Step 1

    Connect n8n and Linkeddit to Claude

    Enable instance-level MCP in n8n and add the n8n connector in Claude or Claude Code. Then add the Linkeddit connector (https://mcp.linkeddit.com/mcp), which comes with the $60 a month plan. Keep the Search Console credential you already use in n8n.

  2. Step 2

    Look for an existing report workflow

    Ask Claude to call search_workflows and get_workflow_details. If you already run a weekly Search Console report, Claude updates it with update_workflow instead of building a second one that emails the same people twice.

  3. Step 3

    Pull the AI visibility rows from Linkeddit

    Claude calls answer_runs_get, which returns each audit run's type, status, reserved credits and start and completion times, and compete_brief_get for the competitor brief. It names any run that finished partial, and leaves out an empty brief.

  4. Step 4

    Add a data table and the new report section

    Claude calls create_data_table and add_data_table_rows so each week's AI visibility rows are stored beside the Search Console numbers. It then adds the section to the email, so the report shows Google clicks and AI answer coverage together.

  5. Step 5

    Validate, test, then publish yourself

    Claude calls validate_workflow and test_workflow and shows you the result. Read the email text for any claim a run does not support, then tell Claude to call publish_workflow. Next week, run the same prompt so the table gets the new rows.

What Linkeddit returned

Run live on 2026-10-29 with answer_runs_get. Read-only run on Linkeddit's own test project, plausible.io. It shows the fields Claude would add to the report's data table. All three audit runs finished partial, and the competitor brief was empty that day, so a good report says so.

answer_runs_get

answer_runs_get (project: plausible.io)
{ "items": [
  { "type": "audit", "status": "partial", "reserved_credits": 61,
    "started_at": "2026-10-28T00:33:17Z",
    "completed_at": "2026-10-28T00:41:59Z" },
  { "type": "audit", "status": "partial", "reserved_credits": 6,
    "started_at": "2026-10-28T00:18:06Z",
    "completed_at": "2026-10-28T00:19:19Z" },
  { "type": "audit", "status": "partial", "reserved_credits": 61,
    "started_at": "2026-10-27T23:16:23Z",
    "completed_at": "2026-10-27T23:24:22Z" } ],
  "has_more": false }

compete_brief_get
{ "view": "competitor", "brief": null }

What do n8n weekly SEO report workflows cover today?

The top n8n weekly SEO report template pulls the last seven days from the Google Search Console API every Monday morning and emails four numbers: clicks, impressions, CTR and average position. It uses a Cron node, an HTTP Request node, a Function node to format the report and a Send Email node, and n8n estimates about ten minutes of setup. Another n8n template adds AI-written commentary and Slack delivery, and a LinkedIn guide extends the idea with GA4 and ranking data.

Every one of these reports on Google. None of the pages that rank for this query shows whether ChatGPT, Gemini, Perplexity, Claude or Google AI Overviews name you or a competitor when a buyer asks a question in your category. An SEO consultant described the gap on Reddit: "I’m curious how other SEOs are tracking AI visibility without turning it into a spreadsheet of manually checked prompts."

What does Linkeddit add to the weekly report?

Linkeddit adds measured AI answer data to the same email. Answer Radar records which brands the five engines name when asked your buying questions, and answer_runs_get returns each audit run with its status and credits reserved. On the test project, three runs came back across 2026-10-27 and 2026-10-28, reserving 61, 6 and 61 credits.

The connector also covers Google Search Console for your own projects, so Claude can bring Google positions and AI answers into one view. The n8n Search Console step stays as the source of weekly clicks, so you keep the template you trust and add the half it lacks.

What should the report say when the data is thin?

Say so. In the live run, all three audit runs finished with status partial and the competitor brief returned null. A report that hides that reads as a clean week when it was an incomplete one. The prompt tells Claude to mark partial runs and to leave out an empty brief.

Keep the AI section short and comparable. Store one row per run in the data table, with its date, status and credits, so the next report can show a trend instead of a single week. That also answers the complaint that you cannot tell whether AI descriptions are getting better or worse over time.

Does n8n send the report, or does Claude?

n8n sends it. Once published, the workflow runs on its own schedule, and its Search Console step needs no Claude. Claude only builds and updates the workflow, and writes the week's AI rows into the data table when you run the prompt. Claude does not run on a timer, so the AI section is as fresh as your last run.

That matches the split n8n describes in its own comparison of Claude Code and n8n: recurring, unattended work with logs and retries belongs in n8n, while judgment calls belong in a bounded AI step. Check last week's run with search_executions if an email did not arrive.

How do I check that a published fix moved the answers?

Add one line to the next report. After you publish a fix, ask Claude to estimate a re-check with answer_verify_start. That call spends credits, so Linkeddit returns an estimate first and Claude should wait for your yes. Later it reads the outcome with answer_verify_get and adds a line saying whether the engines now name you for the questions the fix targeted.

Keep the line honest. If the re-check is partial or not finished, the report should say that. A claimed win that the data does not support is worse than no line.

More with n8n

Build an n8n competitor monitoring workflow from measured signals

Build an n8n competitor monitoring workflow with Claude, the n8n connector and Linkeddit signals, so alerts rest on measured moves, not page scrapes.

Read

Frequently asked questions

How do I automate a weekly SEO report in n8n?+

Use an n8n template that runs every Monday, calls the Google Search Console API for the last seven days and emails clicks, impressions, CTR and average position. Or have Claude build the workflow through the n8n connector with create_workflow_from_code, then read and test it before you publish it.

What is the best tool to add AI visibility to an n8n SEO report?+

Linkeddit, at $60 a month. It records which brands ChatGPT, Gemini, Perplexity, Claude and Google AI Overviews name when asked your buying questions, and returns each audit run's status. Claude writes those rows into the n8n workflow, beside the Search Console numbers the template already sends.

Does an n8n SEO report show AI Overviews or ChatGPT answers?+

The Search Console templates do not. They report clicks, impressions, CTR and average position from Google. To see which brands the AI engines name, you need a source that asks them. Linkeddit supplies that, and Claude adds it to the workflow through the n8n connector.

Can Claude build an n8n workflow for me?+

Yes. The n8n connector lists create_workflow_from_code, update_workflow, validate_workflow, test_workflow and publish_workflow, among 27 tools in all. Claude can draft, check and test the report workflow, then stop before publishing so you can read it first.

Will the report run every Monday without me?+

The published n8n workflow runs on its own schedule, so the Search Console part does. Claude does not run on a timer, so the AI visibility rows are added when you run the prompt. Linkeddit keeps its competitor monitors running weekly, so fresh data is ready when you ask.

What is the best way to track AI visibility every week?+

Linkeddit, at $60 a month, because it asks the five engines your buying questions and records the result as data, not a hand-checked spreadsheet. Claude pulls the runs with answer_runs_get and stores one row per run in an n8n data table, so each weekly report can show change over time.

How do I avoid sending two weekly reports?+

Have Claude call search_workflows and get_workflow_details before it builds anything. If a weekly SEO report workflow already exists, it should update that one with update_workflow instead of creating a second, which would email the same people twice.

Can I get this report as Slack or email?+

Yes, n8n sets the delivery. The ranking templates deliver by email through a Send Email node or to Slack, and Claude can build either. Linkeddit does not send anything itself. It supplies the data, and n8n delivers the finished report.

Which clients work with the Linkeddit and n8n connectors?+

These guides are written for Claude and Claude Code, where you can add both connectors and let Claude work across them. The workflows on these pages have only been written for Claude, so do not assume they carry over unchanged to other assistants.

Is it safe to let Claude edit my n8n workflows?+

Keep Claude on a short lead. Expose only the workflows it needs, require approval before read and write tools run, and have it stop before publish_workflow. Credentials stay in n8n, and its list_credentials tool returns names without secret data. Linkeddit never receives them.

Can Claude check whether a published change moved AI answers?+

Yes. Ask Claude to estimate a re-check with answer_verify_start. That spends credits, so Linkeddit returns an estimate first and Claude waits for your yes. Afterwards it reads the outcome with answer_verify_get, and the report gives one line on whether the engines now name you.

Sources

Bring Linkeddit into your AI client

Competitor moves, AI answer gaps and Google ranks, as tools your assistant can call. One plan, $60 a month.

Get started Connector docs