In-depth architectural comparison of the AgentSEO and Seo MCP MCP servers. Compare execution transports, security boundaries, tool capabilities, quality scores, and ready-to-paste client installation snippets for Claude, Cursor, Windsurf, and VS Code.
At a Glance & Executive Verdict
AgentSEO
Marketing · Local stdio
Quality: 40/100 (Fair) | Auth: No auth required
Seo MCP
Marketing · Local stdio
Quality: 55/100 (Good) | Auth: No auth required
Verdict Summary: Choose AgentSEO if you need specialized Marketing tools running via a local process. Choose Seo MCP if your workspace requires Marketing integration with local subprocess execution. Both servers can be configured concurrently in your client's mcpServers manifest.
Which MCP Server Should You Choose?
Choose AgentSEO when:
You need dedicated capabilities in the Marketing domain.
You prefer local stdio subprocess transport architecture.
Your security boundary fits: No auth required (Free / Open Source).
You have access to required keys: AGENTSEO_API_KEY.
Primary tools included: MCP access to the hosted AgentSEO API, SEO search with query parameters, Location-aware search requests.
Run a one-shot SEO + AI-readability audit on any public URL. Returns scores across 11 modules and ~90 checks, plus actionable findings with rule docs. Limited to 1 audit per IP per 24 hours — for higher volume, get an API key at https://app.metricspot.com/settings/api-keys and use `run_audit`. Synchronous: blocks until the audit completes. Does NOT include Core Web Vitals (use `run_audit` for full PSI scoring). No auth required.
Ready-to-Paste Client Configurations
Paste either (or both) of these JSON server blocks into your client config file (e.g. claude_desktop_config.json or ~/.cursor/mcp.json).
AgentSEO is categorized under Marketing and uses a local stdio subprocess. In contrast, Seo MCP belongs to Marketing using local stdio subprocess. Select AgentSEO when you need capabilities focused on marketing and Seo MCP when you require tools for marketing.
Queue a full SEO + AI-readability audit (includes Core Web Vitals from Google PSI and organic traffic if Google is linked). Returns the audit envelope immediately with `status: queued` and an `audit_id`. Poll `get_audit` with the returned `audit_id` until `status` becomes `complete` (typical 10-30s). Counts against the user's plan allowance. Requires an API key as a Bearer token. Quota and per-domain cooldowns mirror the dashboard.
get_audit
Fetch a previously-run audit by id. Returns module scores (0-100), total score, all findings with severity, recommendation text, and links to the HTML report. Use this to poll a queued `run_audit` until `status: complete`. Requires an API key.
list_audits
List the user's audits (most recent first, deduplicated by URL). Returns `audit_id`, `url`, `status`, `total_score`, `created_at`. Default limit 24, max 100. Use the returned `audit_id` with `get_audit` for full findings. Requires an API key.
get_audit_pdf
Return a signed download URL for the branded PDF report for a given audit id. If no PDF has been rendered yet, queues a render and returns `status: queued` — poll the same tool again, or fetch the URL directly once ready. Requires an API key.
get_organic_traffic
If the user has linked GA4 + Google Search Console, return the 28-day organic traffic snapshot for an audit: session count, daily trend, top landing pages, top queries, and indexing health. Returns `connected: false` if Google is not linked. Cached 24h server-side. Requires an API key.