CitationBench

Agentic SEO Workflow That Scales Your Agency

CitationBench automates the full agentic SEO workflow — research, write, publish, index, and link-build — so a small team can manage 100+ client brands without adding headcount.

Talk to SalesRead the docs
CitationBench · Bootstrap Agent
# Bootstrap a new brand's full SEO presencePOST /v1/agent/invoke> { "agent": "brand-bootstrap", "input": { "url": "https://acme.com" } }job_id: job_9mw4r · status: running# Streaming events via SSE✓ ICP profiles generated (3 segments)✓ Keyword universe built (312 keywords labeled)✓ Content plan ready · awaiting_approval✓ Landing pages queued · 14 pages → WordPress

Fragmented Tools Slow Your Team Down

Running SEO ops across dozens of clients with disconnected tools creates bottlenecks that cost you revenue and client trust.

Onboarding Takes Days Not Minutes

Stitching together Ahrefs, Surfer, Apollo, and AgencyAnalytics for each new client burns hours before any real work begins.

Agents Break Mid-Job and Lose Progress

Fire-and-forget LLM calls have no recovery path — one restart wipes the run and you start over from scratch.

AI Search Is a Blind Spot

No tool in your current stack tells you whether client brands appear in ChatGPT, Claude, or Perplexity responses.

The fix

One Platform. End to End Agentic SEO Ops.

CitationBench composes ~35 tools across research, production, distribution, and measurement into durable agentic workflows. Bootstrap a new brand in 20 minutes, monitor ranks and AI citations daily, and run full link-building campaigns — all from one API or dashboard.

Every Tool Your Agentic Workflow Needs

Five purpose-built capabilities that together cover the full SEO ops lifecycle from first keyword to live outreach campaign.

Run End to End SEO Agents

Workflow Engine

Run End to End SEO Agents

Invoke any named agent via `POST /v1/agent/invoke` with structured input. Supports real-time SSE event streaming, multi-turn continuation, and approval-gated pauses at any step in the workflow.

  • Brand bootstrap in 20 minutes
  • Real-time SSE streaming
  • Approval gates at any step
Pause Agents for Human Review

Human in the Loop

Pause Agents for Human Review

Any agent can emit `awaiting_approval` mid-run. Teams review and approve via dashboard or `POST /v1/agent/invocations/{id}/approve`. Every pause and decision is logged and auditable — safe for client-facing work.

  • Client-safe approval gates
  • Full audit log
  • Resume exactly where stopped
Durable Jobs That Survive Restarts

Reliability

Durable Jobs That Survive Restarts

All agent jobs run on Cyclonic workers — durable, resumable, and cancellable. Cancel via `agent.invocations.cancel`, resume via `agent.invocations.continue`. Never lose a long-running workflow to a server restart.

  • Survives server restarts
  • Cancel and resume anytime
  • Not fire-and-forget
One API Key for Every Client

Agency Architecture

One API Key for Every Client

A single master API key manages all client workspaces. Every call is scoped via the `X-Workspace-Id` header — keywords, content, campaigns, and link-building stay fully isolated per client. White-label-ready.

  • Per-client workspace isolation
  • Bulk actions across clients
  • White-label-ready
Keyword to 50 Outreach Drafts Automatically

Link Building

Keyword to 50 Outreach Drafts Automatically

One `POST /v1/link-building/serp-outreach` call turns a keyword into a SERP scrape, Apollo contact discovery, and 50 personalized email drafts via Instantly. Bulk-approve before sending.

  • 50 drafts from one keyword
  • Apollo partner discovery
  • Bulk approval before send

How it works

From sign-up to first call in five minutes.

01

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.

02

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.

03

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.

04

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 Your Agentic SEO Workflow Today

CitationBench gives your team the full end-to-end agentic SEO workflow — research to link-building — in one platform.

Start Free