MCP for Keyword Research Powered by 2D Intent Data
CitationBench delivers a production-ready hosted MCP server for keyword research with ~35 SEO tools — including industry-first 2D intent × relevance labeling. No self-hosting required.
# Add CitationBench MCP to Claude Code> claude mcp add citationbench --url https://mcp.citationbench.com/mcp✓ Connected · 35 tools available# Run keyword research with 2D intent labelingresearch.keyword.research> { "seed": "mcp for keyword research", "limit": 10, "label": true }{ "kw": "mcp keyword research api", "intent": "COMMERCIAL", "relevance": "CORE" }{ "kw": "seo agent mcp tools", "intent": "PRODUCT", "relevance": "ADJACENT" }# Filter by intent axis before passing to agent> research.keyword.search({ "intent": "COMMERCIAL", "relevance": "CORE" })Keyword Research Tools Weren't Built for Agents
Traditional keyword tools block developers with brittle scraping wrappers, missing intent signals, and months of integration work.
Months of Integration Debt
Stitching Ahrefs exports, SEMrush APIs, and custom scrapers into one agent pipeline takes months and breaks constantly.
No Intent Signal at Scale
Generic keyword APIs return volume and difficulty but no structured intent or relevance axis your agent can filter on.
Self-Hosting Kills Momentum
Running your own MCP server for keyword data means infra overhead before a single SEO workflow goes live.
The fix
One Hosted MCP Endpoint. Every Keyword Tool You Need.
Connect to mcp.citationbench.com/mcp and instantly access keyword research, SERP analysis, content gap detection, and competitor intelligence — all with 2D intent × relevance labels your agent can act on.
Every Keyword Research Tool Your Agent Needs
Five purpose-built MCP tools covering keyword discovery, SERP analysis, intent labeling, competitor intelligence, and content gap detection.
MCP Server
Hosted MCP for Keyword Research Agents
Connect via mcp.citationbench.com/mcp with no self-hosting. ~35 tools span research.*, produce.*, and agent.* namespaces. Works today with Claude Code, Cursor, Claude Desktop, and Windsurf.
- No self-hosting required
- 35 tools via single endpoint
- Claude Code and Cursor ready
Keyword Discovery
Keyword Research via MCP Tool
Call research.keyword.research with a seed URL or phrase. Returns clustered, labeled keyword sets persisted for downstream agent steps. Bulk import via research.keyword.bulk_create for scale.
- Seed URL or phrase input
- Persistent keyword sets
- Bulk import supported
Intent Taxonomy
2D Intent and Relevance Labeling
Every keyword tagged on two axes: intent (informational, commercial, transactional, navigational) and relevance (core, adjacent, tangential). Filter with research.keyword.search so your agent acts on signals, not guesses.
- Intent × relevance axes
- Filter via research.keyword.search
- Bulk relabeling supported
SERP Analysis
Live SERP Fetch for Any Query
research.serp.fetch pulls live SERP results for any keyword, returning top results and domain authority signals. Feed directly into content gap analysis or competitive research steps in your agent pipeline.
- Live SERP per query
- Domain authority signals
- Stored and listable results
Gap Detection
Content Gap Analysis via Agent Workflow
research.content_gap.find compares your keyword inventory against competitor content and target keywords. Returns prioritized gap reports your agent can pass directly into content production.
- Competitor content comparison
- Prioritized gap reports
- Agent pipeline ready
How it works
From sign-up to first call in five minutes.
Sign up and get your key
An `sk_test_*` key lands in your dashboard instantly. No demo gate — start calling real endpoints with shape-complete responses.
Add the MCP server or hit REST
`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.
Run a tool or invoke an agent
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.
Scope to a workspace and ship
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
Built for production, agency-scale, and AI-agent-first.
~35 tools, one MCP server
Hosted at mcp.citationbench.com/mcp. Works with Claude Code, Cursor, Claude Desktop, Windsurf, and ChatGPT Apps. No self-hosting required.
Durable jobs, not fire-and-forget
Every long-running call runs on Cyclonic workers — survives restarts, cancellable, resumable, streams via SSE. Production-safe by default.
Multi-workspace from day one
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.
Demo mode out of the box
Every endpoint responds in shape-complete demo mode without auth, so you can build the agent before the user signs up.
FAQ
Common questions
Start Using MCP for Keyword Research Today
Connect your agent to a hosted MCP server with ~35 SEO tools and 2D intent labeling — no infrastructure required.
View MCP Docs