CitationBench provides a hosted MCP server for SEO with ~35 tools — connect Claude Code, Cursor, or any MCP client and skip months of API integration work.
# Add CitationBench to Claude Code> claude mcp add citationbench --url https://mcp.citationbench.com/mcp✓ Connected · 35 tools available# Fetch live SERP and analyze keyword opportunityresearch.serp.fetch> { "query": "mcp server for seo", "workspace_id": "ws_acme" }{ "results": 10, "cliff_detected": true, "cliff_position": 6, ...# Run keyword research with 2D labelingresearch.keyword.research> { "seed": "mcp seo tools", "limit": 30, "label": true }{ "keywords": [{ "kw": "mcp server seo", "intent": "COMMERCIAL", ...Wiring LLMs to SEO data sources means months of integration work before a single agent runs reliably.
Connecting Claude to DataForSEO, SE Ranking, and GSC separately means building and maintaining every API client yourself.
Fire-and-forget LLM calls have no approval gates, no resume-on-failure, and no audit trail — production agents need more.
Discovering what SEO operations are even possible requires reading docs across a dozen vendors before writing a single line of agent logic.
The fix
CitationBench's hosted MCP server at mcp.citationbench.com/mcp exposes ~35 tools across research, production, distribution, and agents namespaces. Connect from Claude Code, Cursor, or Windsurf in one command and invoke keyword research, SERP analysis, rank tracking, link building, and AI citation monitoring from any agent.
Five capability areas — covering the full SEO workflow — available as named MCP tools with no self-hosting required.
MCP Infrastructure
Production-ready server at mcp.citationbench.com/mcp requires no self-hosting. Tools namespaced as research.*, produce.*, indexing.*, link_building.*, and agent.* — callable from Claude Code, Cursor, Claude Desktop, Windsurf, and ChatGPT Apps.
Research Tools
Invoke research.keyword.research from your agent to discover, cluster, and 2D-label keyword sets from a seed. Results persist for downstream production steps. Bulk import via research.keyword.bulk_create.
SERP Analysis
Call research.serp.fetch to pull live SERP data for any query directly inside an agent workflow. Returns top results, domain signals, and content types — ready to pass to content generation or gap analysis tools.
Agent Orchestration
Use agent.invoke to run named agents like brand-bootstrap directly from your MCP client. Supports real-time SSE event streaming, multi-turn continuation, and approval-gated pauses — not fire-and-forget calls.
Measurement
Pull daily rank data per keyword across all client workspaces from your agent. Scoped via X-Workspace-Id header — no separate tracker to wire up. Available via MCP tool, REST, and dashboard simultaneously.
How it works
An `sk_test_*` key lands in your dashboard instantly. No demo gate — start calling real endpoints with shape-complete responses.
`claude mcp add citationbench https://mcp.citationbench.com/mcp` from Claude Code, Cursor, or any MCP client. Or `curl` against `api.citationbench.com/v1/*` directly.
Every tool returns shape-complete demo data without auth, so your agent works before the user signs up. Tools across research, production, indexing, link-building, and agents.
Add `X-Workspace-Id: ws_***` to scope per client. Switch to a live `sk_live_*` key when you're ready. Same API surface, same SDK.
Why CitationBench
Hosted at mcp.citationbench.com/mcp. Works with Claude Code, Cursor, Claude Desktop, Windsurf, and ChatGPT Apps. No self-hosting required.
Every long-running call runs on Cyclonic workers — survives restarts, cancellable, resumable, streams via SSE. Production-safe by default.
One master API key, N client workspaces. Switch with a single `X-Workspace-Id` header. Per-client data isolation, bulk ops across all of them.
Every endpoint responds in shape-complete demo mode without auth, so you can build the agent before the user signs up.
FAQ
CitationBench's hosted MCP server gives your AI agents a complete SEO tool layer with no self-hosting or integration work.
View MCP Docs