Manual approval processes delay campaigns and frustrate clients. CitationBench automates end-to-end SEO workflows with built-in human approval gates that pause, log, and resume at any step.
Disconnected tools and ad-hoc approval chains erode client trust and stall campaign execution across your portfolio.
Without structured approval gates, unapproved content goes live and damages client relationships you spent months building.
Email threads and Slack messages replace a real system, burying critical client feedback and pushing delivery dates back.
Generic project tools lack logged approval decisions, leaving you with no record when a client questions what was authorized.
The fix
CitationBench agents pause at configurable checkpoints, emit an awaiting_approval status, and wait for human sign-off via the dashboard or API before proceeding. Every pause, review, and decision is logged and auditable across all client workspaces.
CitationBench combines durable agent jobs, structured approval gates, and multi-workspace isolation so agencies can run SEO ops across every client brand without losing control.
Approval Gates
Any agent pauses and emits awaiting_approval at any configured step. Clients or internal reviewers approve or reject via dashboard or POST /v1/agent/invocations/{id}/approve. The agent resumes exactly where it stopped.
Workflow Agent
POST /v1/agent/invoke runs named agents across research, production, and distribution with real-time SSE streaming. Approval-gated pauses prevent any action from executing without explicit sign-off.
Agency Architecture
A single master API key scopes all SEO ops, approval chains, and content per client via the X-Workspace-Id header. Approvals in one workspace never bleed into another.
Job Reliability
Jobs are backed by durable Cyclonic workers — not fire-and-forget LLM calls. A workflow paused for client approval stays paused and resumes cleanly via agent.invocations.continue, even after a server restart.
Notifications
Register per-workspace webhooks that fire on agent.invocation.awaiting_approval and produce.publish.completed events. Wire approval requests directly into Slack, email, or your client portal without polling.
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 gives SEO agencies structured, auditable approval gates built into every workflow so client sign-off never blocks your team again.
Book a Demo