Skip to main content
Cituna
AI Visibility

The AI visibility API

A read-and-write REST API for AI-visibility data across seven engines, ChatGPT, Perplexity, Gemini, Claude, Grok, Google AI Overviews and Google AI Mode. JSON over HTTPS, bearer-token auth, included from the $39 Starter plan.

By Rahul AUpdated September 12, 20267 min read

See which of these you are already failing.

On this page
  1. What the API does
  2. Authentication
  3. Endpoints
  4. Example request
  5. Prefer no code? Use MCP
  6. Plans and rate limits
  7. FAQ

Most AI-visibility tools give you a dashboard and stop there. Cituna gives you the data. Every number you see in the app, which of the seven AI engines cited your brand, at what position, for which buyer question, and what each engine actually wrote, is available over a plain JSON REST API you can pull into your own dashboard, BI tool, agent or CI pipeline.

What the API does

The API returns structured AI-visibility data across seven engines, ChatGPT, Perplexity, Gemini, Claude, Grok, Google AI Overviews and Google AI Mode (we do not track Microsoft Copilot). You can read your daily tracking grid, the receipts behind it, your audits and gap queue, your citation sources and your live Google Search Console metrics, and you can write back: start a scan, move a gap along your fix queue, or queue an article for AutoSEO.

Authentication

Generate a personal API key in the app under Integrations, then “Claude / MCP access”, then Generate token. The key looks like cituna_sk_… and is shown once. Send it on every request as a bearer token:

Authorization: Bearer cituna_sk_your_token

The same key authenticates the REST API and the MCP server. Keys are long-lived, scoped to your own workspace, and revocable from the app at any time. Cituna never receives your password or your Google credentials.

Endpoints

Base URL https://cituna.com/api. All responses are JSON. Read endpoints work from the $39 Starter plan; the write endpoints require Pro or higher, the same split as the MCP write tools.

Method and pathWhat it returnsPlan
GET /auth/meYour account, workspace, plan, and the active scoring epoch and engine count.From $39
GET /projectsThe brands and domains tracked in your workspace, with their competitors and prompts.From $39
GET /visibility?domain=Today’s tracking grid: per prompt and per engine, whether you were cited and at what position.From $39
GET /engine-answers?domain=&prompt=The receipts: what each engine answered for a prompt, the brands it named and the URLs it cited.From $39
GET /history/:domainYour AI-visibility audits over time, scores and gap counts.From $39
GET /sources?domain=The third-party hosts engines cite for your category, and whether your brand is on them.From $39
GET /seo/snapshots?domain=Authority and SEO snapshots for a domain.From $39
GET /me/usageYour per-tool usage meters and remaining allowance.From $39
GET /integrations/gsc/statusWhether Google Search Console is connected and which properties are available.From $39
GET /integrations/gsc/data?domain=Search Console clicks, impressions, CTR and position for a domain.From $39
POST /integrations/gsc/queryRun a Search Analytics query with your own dimensions, filters and date range.From $39
POST /scanStart a fresh AI-visibility scan for a URL (Server-Sent-Events stream; uses one scan from your quota).Pro
POST /gap-statusMove a gap along your fix queue (todo / doing / done).Pro
POST /autoseo/queueQueue a keyword so AutoSEO drafts the page on its next run.Pro
POST /autoseo/publishedTell Cituna a page for a keyword is live, so it stops suggesting the topic.Pro
POST /sources/outreachUpdate the outreach status of a citation-source target.Pro

Base URL https://cituna.com/api. Authenticate every call with Authorization: Bearer cituna_sk_…. Read endpoints work on Starter and up; write endpoints (scan, gap-status, autoseo/queue, autoseo/published, sources/outreach) require Pro or higher. Verified live on September 12, 2026.

Example request

Read your daily tracking grid for one domain:

curl https://cituna.com/api/visibility?domain=acme.com \
  -H "Authorization: Bearer cituna_sk_your_token"

Returns the per-prompt, per-engine grid (abridged):

{
  "found": true,
  "brand": { "domain": "acme.com", "name": "Acme" },
  "score": { "value": 42, "label": "Growing" },
  "measuredDay": "2026-09-12",
  "engines": ["chatgpt", "perplexity", "gemini",
              "claude", "grok", "aioverviews", "aimode"],
  "prompts": [
    {
      "prompt": "best project management tool",
      "cited_by": 3,
      "tested_by": 7,
      "cells": {
        "chatgpt":    { "cited": true,  "position": 2, "status": "cited" },
        "perplexity": { "cited": false, "status": "answered" }
      }
    }
  ]
}

The engines array is the canonical order of the seven engines; each prompt’s cells object is keyed by that same engine id (aioverviews is Google AI Overviews, aimode is Google AI Mode). cited_by is how many of the seven engines cited you for that prompt; tested_by is how many ran.

Prefer no code? Use MCP

If you want your AI-visibility data inside Claude (or any MCP client) rather than in your own code, skip the REST calls and add the Cituna MCP server. It wraps these same endpoints as 16 tools, connects in three steps with nothing installed, and uses the same cituna_sk_ token. The REST API is the right choice when you are building a dashboard, a scheduled job or your own product on top of the data.

Same data, two front doors:

The MCP server and the REST API read and write the same workspace through the same key and the same plan rules. Use MCP for conversational, no-code access; use REST when you need to script it.

Plans and rate limits

API access is included from Starter ($39/mo), with read endpoints available straight away. Pro ($119/mo) adds the write endpoints and a larger monthly allowance; Max ($399/mo) raises the scan, data and API allowances and adds priority support; a custom Enterprise plan lifts them further with invoiced billing. Calls meter against your monthly allowance (see GET /me/usage); scans meter separately against your scan quota, because each one runs live model queries. There is no per-call charge and no separate API add-on. See the full pricing for the per-plan limits.

Drafted with AI assistance from our own research and Search Console data, and reviewed by Rahul A before publishing. Rules and prices change; check the linked official source before you act.

Frequently asked questions

Is there an AI visibility tool with a REST API for developers?

Yes. Cituna exposes a JSON REST API at https://cituna.com/api, authenticated with a bearer token (Authorization: Bearer cituna_sk_…). It returns the same data the dashboard shows: per-engine brand citations and positions across seven AI engines, the raw engine answers, AI-visibility audits, citation sources and your Google Search Console numbers. It is included from the $39 Starter plan, so you do not need an enterprise contract for programmatic access.

Which AI engines does the API cover?

Seven: ChatGPT (OpenAI), Perplexity, Gemini (Google), Claude (Anthropic), Grok (xAI), Google AI Overviews and Google AI Mode. Every response that breaks visibility down by engine returns all seven. Microsoft Copilot is not covered, because no data source reads it.

How do I authenticate with the Cituna API?

Generate a personal API key in the app under Integrations, then "Claude / MCP access", then Generate token. The key looks like cituna_sk_… and is shown once. Send it as an HTTP header: Authorization: Bearer cituna_sk_your_token. The same key works for the REST API and the MCP server. Keys are long-lived and revocable; revoke one anytime from the app.

Does the Cituna API have an OpenAPI spec or SDK?

The fastest integration is the MCP server (npx -y cituna-mcp), which wraps the same API and works with any MCP client with no code. For direct REST use, the endpoints are documented on this page and return plain JSON, so any HTTP client works. A formal OpenAPI document and language SDKs are on the roadmap; if you need the schema before then, ask us through the Enterprise contact.

What are the Cituna API rate limits?

Calls are metered per plan against your monthly MCP and API allowance, visible at GET /api/me/usage. Scans are metered separately against your scan quota, because each scan runs live model queries. Starter includes API access with the Starter allowance; Pro and Max raise it, and Max adds priority support. There is no separate API add-on and no per-call charge.

Can the API run scans and change my data, or is it read-only?

Both, scoped to your plan. Read endpoints (visibility, engine answers, audits, sources, Search Console) work from $39 Starter. Write endpoints, start a scan, move a gap along your fix queue, queue or publish an AutoSEO article, update outreach status, need Pro or higher, exactly like the MCP write tools. A key acts only within your own workspace.

See how AI engines see your brand

Start a free 3-day trial and see the exact buyer prompts you lose across ChatGPT, Perplexity, Gemini, Claude, Grok, Google AI Overviews and Google AI Mode, with a prioritized AEO, GEO and SEO action plan and the fixes to win them.

3-day free trial · Card required, cancel anytime · Works with ChatGPT, Perplexity, Gemini, Claude, Grok, Google AI Overviews and Google AI Mode

Start free trial