# OpenSEO

> Self-hostable SEO workbench that also serves its data to agents over MCP

- Category: AI Productivity & Writing / SEO & AI Search Visibility
- Language: TypeScript
- Licence: MIT
- Repository: https://github.com/every-app/open-seo
- Website: https://openseo.so
- Canonical page: https://ai-tldr.dev/tools/open-seo/

## In short

OpenSEO is an open-source SEO platform that covers the work the big subscription suites do — keyword research, rank tracking, competitor insight, backlink analysis, site audits and AI-visibility checks — but changes the commercial shape. It sits in the SEO & AI Search Visibility category of the AI Productivity & Writing tools and is open-source, written in TypeScript and licensed MIT.

## Overview

OpenSEO is an open-source SEO platform that covers the work the big subscription suites do — keyword research, rank tracking, competitor insight, backlink analysis, site audits and AI-visibility checks — but changes the commercial shape. The README calls it "an SEO tool for the people": you bring your own DataForSEO API key and pay DataForSEO directly for the data you pull, instead of paying a monthly seat price to a vendor sitting in between.

The AI angle is not a chat box bolted on the side. OpenSEO exposes an MCP server, so agents such as Claude Code, OpenClaw and Hermes can query your SEO data directly — pull the keyword set, read the audit findings, check where a page ranks — and act on it in the same session where they are editing the site. Reusable workflows are packaged as Agent Skills.

It is MIT-licensed and written mainly in TypeScript, and the repository documents three ways to run it: local development, self-hosting with Docker, and self-hosting on Cloudflare. A hosted instance is also available at openseo.so for anyone who would rather not operate it.

## What it does

- Keyword research, rank tracking and competitor insight
- Backlink analysis and site audits
- AI-visibility tooling for tracking how a site surfaces in AI answers
- MCP server so agents like Claude Code, OpenClaw and Hermes can query your SEO data directly
- Agent Skills packaging reusable SEO workflows
- Bring-your-own DataForSEO key — pay the data provider directly, with no subscription in between
- Self-hostable with Docker or on Cloudflare

## Getting started

OpenSEO needs a DataForSEO API key for its data. The repository documents local development plus two self-hosting paths; pick the one that matches where you want it to run.

1. Get the source

   Clone the repository — the setup guides live in its docs directory.

```bash
git clone https://github.com/every-app/open-seo.git
cd open-seo
```

2. Follow the setup guide for your target

   Local development is documented in docs/LOCAL_DEVELOPMENT.md, Docker self-hosting in docs/SELF_HOSTING_DOCKER.md, and the Cloudflare deployment in docs/SELF_HOSTING_CLOUDFLARE.md.

3. Add your DataForSEO key

   OpenSEO does not resell data: create a DataForSEO account, add the key to your environment, and you are billed by DataForSEO for the queries you run.

4. Wire it into an agent

   Point an MCP-capable agent at the OpenSEO MCP server to query your data from a coding or writing session. Individual skills can also be installed on their own.

```bash
npx skills add every-app/open-seo --skill simple-issue-description
```

## When to use it

- Reach for it when Semrush or Ahrefs seat pricing is out of proportion to how much SEO data you actually pull
- Use it when you want an agent to read audit findings and rankings while it edits the site, rather than copying numbers out of a dashboard
- Good for agencies or teams that need the SEO stack self-hosted on their own infrastructure
- Useful for tracking AI visibility alongside classic keyword and backlink work