CitationBench is a hosted MCP server for content marketing with ~35 SEO and GEO tools — callable from Claude Code, Cursor, or any MCP client without self-hosting.
# Add CitationBench MCP to Claude Code> claude mcp add citationbench --url https://mcp.citationbench.com/mcp✓ Connected · 35 tools available# Research content gaps for a client domainresearch.content_gap.find> { "domain": "acme.com", "competitors": ["rival.com"], "limit": 10 }{ "gaps": [{ "topic": "ai seo tools comparison", "priority": "high" }, ...# Generate and publish a blog post from the gapproduce.blog_post.create> { "keyword": "ai seo tools comparison", "workspace_id": "ws_acme" }✓ Draft ready · status: awaiting_approvalStitching together separate tools for research, writing, publishing, and tracking burns engineering hours and breaks workflows.
Connecting Ahrefs, Surfer, and Apollo via separate APIs takes months before a single piece of content ships.
Generic MCP servers expose raw search or writing tools but cannot run a full research-to-publish-to-index workflow in one agent.
Fire-and-forget LLM calls fail silently on multi-step content workflows with no way to resume or audit what happened.
The fix
CitationBench exposes ~35 purpose-built SEO and GEO tools through a single hosted MCP server at mcp.citationbench.com/mcp — covering keyword research, content gap analysis, blog generation, multi-platform publishing, auto-indexing, and rank tracking in one callable tool layer.
Five feature areas covering the full content marketing workflow — research, production, distribution, orchestration, and multi-client management — all callable as MCP tools.
MCP Infrastructure
Connect to mcp.citationbench.com/mcp in one command. All ~35 tools namespaced across research.*, produce.*, indexing.*, and link_building.* — no self-hosting, no infra work.
Content Research
research.content_gap.find compares your client's content inventory against competitor content and target keywords, returning prioritized gap reports your agent can act on immediately.
Content Production
produce.blog_post.create and produce.blog_post.bulk_create turn keyword lists into full SEO drafts. Section-level regeneration and brief-to-draft pipelines keep output quality consistent.
Publishing
Every produce.publish call automatically fires Google Search Console URL submission and IndexNow simultaneously. No manual indexing step — content is discoverable from the moment it goes live.
Agent Control
Agents pause and emit awaiting_approval at any step. Reviewers approve or reject via dashboard or POST /v1/agent/invocations/{id}/approve. Every decision is logged and auditable — not fire-and-forget.
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 to your MCP client and run the full content marketing workflow from research to publish in one agent.
View MCP Docs