Stop building SEO integrations from scratch. Connect your Claude Code agent to CitationBench's hosted MCP server and get 35+ tools for research, production, and distribution.
# Add CitationBench as your Claude Code SEO skill> claude mcp add citationbench --url https://mcp.citationbench.com/mcp✓ Connected · 35 tools available# Research keywords with 2D intent × relevance labelingresearch.keyword.research> { "seed": "seo automation tools", "limit": 25, "label": true }{ "kw": "agentic seo platform", "intent": "COMMERCIAL", "relevance": "CORE" }# Invoke a brand bootstrap agent from your agentagent.invoke> { "agent": "brand-bootstrap", "input": { "url": "https://acme.com" } }job_id: job_4zt9m · status: running · awaiting_approvalStitching together raw APIs costs months of integration work your clients cannot afford.
Wiring Ahrefs, Surfer, Apollo, and Instantly into a Claude Code agent takes months, not days.
Claude Code has no native SEO tooling, leaving every research and publishing step to custom code.
LLM calls without durable jobs and approval gates break silently and leave clients with no audit trail.
The fix
CitationBench's hosted MCP server at mcp.citationbench.com/mcp gives Claude Code instant access to 35+ tools across research, content production, publishing, indexing, and link-building — with durable jobs and human approval gates.
Select from 35+ MCP tools across namespaces covering the full organic-visibility workflow from research through distribution.
MCP Integration
Connect Claude Code to CitationBench with one command. The hosted server at mcp.citationbench.com/mcp exposes ~35 tools across research.*, produce.*, indexing.*, link_building.*, and agent.* namespaces. No self-hosting required.
Research
Call research.keyword.research from your Claude Code agent to discover, cluster, and label keyword sets from a seed URL or list. Results persist in CitationBench for downstream production workflows.
Orchestration
Use agent.invoke to run named agents like brand-bootstrap directly from your Claude Code session. Supports multi-turn continuation, real-time SSE streaming, and approval-gated pauses at any step.
Compliance
Any agent running via Claude Code can pause and emit awaiting_approval before a sensitive action. Approve or reject via dashboard or POST /v1/agent/invocations/{id}/approve. Every decision is logged and auditable.
Reliability
All agent jobs run on CitationBench's Cyclonic worker system — durable, resumable, and cancellable. Cancel via agent.invocations.cancel or resume via agent.invocations.continue. Not fire-and-forget LLM calls.
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
Connect CitationBench's hosted MCP server to Claude Code and access 35+ SEO tools with no integration work.
View MCP Docs