Open-SEO Main Features: Technical Deep-Dive into the Open-Source SEO Platform

Open-SEO is a full-stack, open-source SEO platform providing keyword research, rank tracking, competitor insights, backlink analysis, site audits, and AI agent integration through a pay-as-you-go credit system designed as an alternative to expensive commercial tools like Semrush and Ahrefs.

The every-app/open-seo repository delivers these capabilities via a modern TypeScript architecture built on TanStack React-Start and Cloudflare Workers. It integrates directly with Google Search Console and DataForSEO APIs while exposing functionality through both a focused React UI and a Model-Context-Protocol (MCP) server for AI agents.

Core SEO Workflow Features

Open-SEO implements six primary SEO workflows through typed server functions and dedicated UI routes defined in web/src/lib/feature-pages.ts.

Keyword Research

The keyword research module generates keyword ideas, difficulty scores, and topic clusters by querying Google Search Console data. The implementation in src/serverFunctions/searchPerformance.ts builds GSC filters and returns totals, striking-distance rows, and country lists.

import { getSearchPerformanceReport } from "@/serverFunctions/searchPerformance";

await getSearchPerformanceReport({
  method: "POST",
  data: {
    dateRange: "last-30-days",
    device: "mobile",
    country: "US",
  },
});

This function calls GscService.getPerformance and powers the UI routes under /features/keyword-research.

Rank Tracking

The rank tracking feature monitors position changes for selected keywords across time. Implemented in src/serverFunctions/rank-tracking.ts, it stores historical SERP data and exposes trend analysis through the /features/rank-tracker interface.

Competitor Insights

Competitor analysis leverages src/serverFunctions/serp-locations.ts to retrieve SERP positioning data and backlink profiles for competing domains. The UI routes under /features/competitor-insights visualize this comparative data.

Backlink Analysis

The backlink checker retrieves and monitors backlink profiles via DataForSEO integration. Server functions in src/serverFunctions/backlinks.ts (referenced in the architecture) support the /features/backlink-checker workflow.

Site Audits

Site auditing crawls websites to detect broken links, duplicate content, and missing meta tags. The crawl workflow resides in src/server/workflows/site-audit-workflow-helpers.ts, while MCP exposure lives in src/server/mcp/tools/site-audit-tools.ts.

import { mcpResponse } from "@/server/mcp/formatters";
import { withMcpProjectAuth } from "@/server/mcp/project-auth";

export const runSiteAudit = withMcpProjectAuth(
  async (projectId, url) => {
    await startCrawl({ projectId, startUrl: url });
    return mcpResponse({ status: "queued", url });
  }
);

AI Visibility

The AI visibility layer exposes all SEO data through MCP tools (e.g., whoami.ts, search-console-tools.ts) allowing external AI agents to query statistics, trigger audits, and receive remediation instructions programmatically.

MCP Server and AI Integration Architecture

Open-SEO's MCP server functions as a bridge between SEO data and AI agents like Claude Code or Hermes. The transport layer in src/server/mcp/transport.ts handles authentication, versioning, and response formatting.

Individual tools reside in src/server/mcp/tools/*:

• whoami.ts – Minimal example demonstrating project authentication
• search-console-tools.ts – Exposes GSC performance statistics
• site-audit-tools.ts – Provides audit triggers and issue remediation

These typed tools enable AI skills (reusable workflows) defined in docs/MAINTAINERS.md that guide agents through complex SEO tasks without manual UI interaction.

Technical Implementation Stack

Front-End Architecture

The React-Start application provides a modern interface with file-based routing managed by web/src/routeTree.gen.ts. Feature pages are registered in web/src/lib/feature-pages.ts with associated metadata and keywords.

Server Functions and Data Layer

TanStack React-Start server functions (located in src/serverFunctions/) call third-party services and return typed results. The data layer uses D1 (SQLite-compatible) for structured storage and optionally R2 for object storage, with schemas defined via Drizzle ORM in drizzle.config.ts and src/db/telemetry.schema.ts.

Self-Hosting Options

Open-SEO supports two officially documented deployment paths:

1. Docker – Local deployment using Dockerfile.selfhost and documented in docs/SELF_HOSTING_DOCKER.md
2. Cloudflare Workers – Edge-distributed deployment using wrangler.jsonc for high-throughput crawls within the Cloudflare step budget, documented in docs/SELF_HOSTING_CLOUDFLARE.md

Both paths share the same core codebase, differing only in deployment scripts.

Credit-Based Billing System

Open-SEO operates on a credit-based pricing model defined in src/shared/billing-credit-features.ts. The hosted SaaS charges $10 per month, while self-hosted deployments only incur DataForSEO API usage costs without platform fees.

import { getProjectCredits } from "@/shared/billing";

const credits = await getProjectCredits({ projectId: "proj_123" });
console.log(`Remaining credits: ${credits}`);

This implementation allows granular cost control compared to traditional subscription-based SEO tools.

Summary

• Open-SEO provides six core SEO workflows: keyword research, rank tracking, competitor insights, backlink analysis, site audits, and AI visibility
• The MCP server architecture in src/server/mcp/transport.ts enables AI agent integration through typed tools like search-console-tools.ts and site-audit-tools.ts
• Technical stack includes TanStack React-Start, Cloudflare Workers/D1, and Drizzle ORM for type-safe database operations
• Self-hosting supports both Docker local deployment and Cloudflare Workers edge distribution
• Billing uses a credit system (src/shared/billing-credit-features.ts) with $10/month hosted option or pay-for-usage DataForSEO costs when self-hosted

Frequently Asked Questions

What makes Open-SEO different from Semrush or Ahrefs?

Open-SEO is open-source and offers a pay-as-you-go credit system rather than expensive fixed subscriptions. According to the every-app/open-seo source code, you can self-host the platform and only pay for DataForSEO API usage, or use the hosted version for $10/month, significantly reducing costs for agencies and individual practitioners.

How does the MCP server integration work?

The MCP server exposes SEO data through typed tools in src/server/mcp/tools/* that AI agents can invoke. The transport layer in src/server/mcp/transport.ts handles authentication and formatting, allowing tools like search-console-tools.ts to return GSC data directly to agents like Claude Code for automated analysis and recommendations.

Can I self-host Open-SEO without using Cloudflare?

Yes. While Cloudflare Workers provide scalable edge deployment documented in docs/SELF_HOSTING_CLOUDFLARE.md, the repository includes docs/SELF_HOSTING_DOCKER.md and a Dockerfile.selfhost for local Docker-based deployments that run independently of Cloudflare's infrastructure.

What database does Open-SEO use for storing SEO data?

Open-SEO uses D1, a SQLite-compatible database, for structured data storage with schemas defined in Drizzle ORM files like drizzle.config.ts and src/db/telemetry.schema.ts. It optionally uses R2 for object storage, particularly for crawl data and large audit reports.

{
  "mcpServers": {
    "instagit": {
      "command": "npx",
      "args": ["-y", "instagit@latest"]
    }
  }
}