Add https://mcp.omnifamous.com to your AI client and sign in when it asks: pick your workspace and Editor or Viewer access. Clients without browser sign-in (n8n, Zapier, scripts, CI) use a workspace secret key from MCP & API settings, sent as an Authorization: Bearer header.

What costs no credits?

Static page inspection, robots/sitemap checks, cached AI visibility snapshots and the engine list cost no credits once you sign in, subject to per workspace rate limits. There is no automatic signup-credit grant.

One workspace balance is shared by your team and agents. 2,000 credits equal $1 of usage value. Metered tools reserve their quoted price and charge on a successful result; failed calls release the reservation. Purchased and plan-granted credits roll over.

Can credits refill automatically?

A workspace admin can explicitly save a card and enable automatic refill, with a low-balance threshold, refill amount and monthly limit on automatic card charges. It starts off. Low balances are checked every five minutes.

Does this charge again for my subscription?

Scheduled work included in Content Writer and AI Visibility stays covered by those subscriptions. Workspace visibility reads have no per-call credit charge and require AI Visibility access. Buying credits alone does not unlock that subscription.

Can the agent draft or publish articles through this server?

This release exposes research, page checks and existing visibility reports. Article generation and publishing are not in the public MCP catalog.

Can I run an arbitrary AI prompt?

The domain visibility check generates its own prompts. The workspace tools read results already captured by monitoring. Custom prompt execution is not in this release.

Can I give an agent unlimited access?

Use a revocable workspace key and keep it private. Requests are rate limited and metered work is bounded by the workspace's credit balance and spend limits. Agent keys cannot enable automatic payments or change billing settings.

Waitlist

$ omnifamous seo mcp live

The SEO MCP server your agent keeps

Your search data, in the conversation.

An SEO MCP server connects your AI agent to keyword research, traffic estimates, backlink data and AI visibility reports. Connect your workspace once, ask in plain language, and get structured data your agent can turn into a useful analysis.

See client configs

36 tools one workspace balance four zero-credit tools

claude mcp add --transport http omnifamous https://mcp.omnifamous.com

// works with your agent

Pick your agent. Each one has its own setup page.

Choose your client for connection instructions. Most clients only need the server URL and a browser sign-in; headless ones use a workspace API key.

Claude CodeGet started CodexGet started CursorGet started ClaudeGet started ChatGPTGet started n8nGet started ZapierGet started Gemini CLIGet started VS CodeGet started GitHub CopilotGet started WindsurfGet started ClineGet started HermesGet started JetBrainsGet started ZedGet started WarpGet started

// what you can build

Useful questions. Evidence your agent can work with.

Research a domain, prioritize keywords or understand your existing monitoring results. Pick a workflow and give your agent the question.

Understand your existing visibility results

Ask your agent to explain how visibility has changed, which engines mention you, and where citations concentrate. These reads use stored monitoring data and require AI Visibility access.

yourdomain.com/ai-visibilitybuilt by your agent

Illustrative layout with sample data. Your agent receives the tool results described below.

>Summarize my workspace's AI visibility over the last 30 days. Explain the coverage, trends and most-cited domains.

⏺ ai_visibility_get_summary⏺ ai_visibility_get_trend⏺ ai_visibility_top_cited_domains⏺ A summary tied to your existing monitoring evidence.

// a domain check

See whether a brand appears in AI answers.

The domain visibility tool asks a set of brand-related questions and returns presence results. It is a snapshot. Your workspace monitoring reports provide the existing history and coverage context.

claude, connected to omnifamousSep 20, 2026 example, response trimmed

> Do ChatGPT, Claude and Gemini actually know linear.app? Check it.

⏺ ai_visibility_run_scan (domain: linear.app)

{
  "brand": "Linear",
  "domain": "linear.app",
  "prompts": [
    "Tell me about Linear. What does the company do?",
    "What are the best alternatives to Linear?",
    "Is Linear a good choice? Share any reviews."
  ],
  "overallPresence": 100,
  "platforms": [
    { "label": "ChatGPT", "appeared": true, "mentions": 3, "totalPrompts": 3 },
    { "label": "Claude",  "appeared": true, "mentions": 3, "totalPrompts": 3 },
    { "label": "Gemini",  "appeared": true, "mentions": 3, "totalPrompts": 3 }
  ]
}

⏺ Linear is named in all three answers on every prompt. Want me to run the same check on your domain, so you can see where the gap is?

What is an SEO MCP server?

MCP is the open protocol that lets an AI agent call external tools. An SEO MCP server is that protocol pointed at search data: your agent asks for keyword metrics, a backlink summary or existing visibility results and receives structured data it can analyze. Here is the connection flow:

Add the server URL

Paste https://mcp.omnifamous.com into your AI client, or copy its config below. ChatGPT, Claude, Claude Code, Cursor, VS Code and most other clients need nothing else.

Sign in and pick a workspace

Your client opens a browser. Sign in, choose the workspace and Editor or Viewer access. Editor can spend credits; Viewer covers the free tools and workspace reads. Headless clients use a workspace API key instead.

Ask in plain language

Compare competitors, find keyword opportunities, or summarize your existing AI visibility reports. Your agent selects the tools.

See what was used

Metered calls draw from your workspace credits. Check the balance and recent charges in Billing, and opt into automatic refill if you need it.

// the catalog

Every tool, with its real signature

All 36 tools, with their inputs and prices. Saved dashboard reads require AI Visibility access, Search Console tools require a connected property, and monitors are paid per run. Every tool is rate limited, including zero-credit reads.

Keyword research

Explore seed keywords with provider-reported volume, difficulty, CPC and intent.

Traffic estimates

Compare estimated organic traffic, ranked keywords and position distributions.

Competitor gaps

Find keywords a competitor ranks for that your domain is missing.

Backlink summaries

Compare backlinks, referring domains and authority signals for a domain.

AI visibility reports

Run a domain visibility check or read your existing workspace's monitoring results.

Page checks

Inspect static HTML, robots.txt, sitemap files and up to 40 outbound page links.

Zero credits

tool	arguments	returns	cost
`audit_page`	`url: string`	static HTML tags, final status, canonical, robots meta; no JavaScript rendering	free
`audit_robots_and_sitemap`	`url: string`	robots.txt rules and counts, sitemap index, page count, sample URLs	free
`ai_visibility_get_snapshot`	`domain: string`	the last 12 hours' cached AI visibility scan per engine, or not_scanned with the free report link; never runs a scan	free
`ai_visibility_list_engines`	`surface?: app \| api \| search_api`	every engine we measure, grouped as apps captured like a user, model APIs (with or without web search) and search APIs, with model, region meaning and credit multiplier	free

Keywords and traffic

tool	arguments	returns	cost
`find_keywords`	`seed: string, countryCode?: string`	keyword, volume, difficulty, cpc, intent, long tail	400 credits ($0.20)
`organic_research`	`target: string, countryCode?: string`	estimated traffic, organic keywords, traffic value, position bands	400 credits ($0.20)

Links and competitors

tool	arguments	returns	cost
`backlinks_research`	`target: string`	referring domains, backlinks, authority, broken backlinks, dofollow split	600 credits ($0.30)
`competitors_research`	`yourDomain: string, competitorDomain: string, countryCode?: string`	keyword gaps, shared keywords, traffic opportunity	1,800 credits ($0.90)
`audit_broken_links`	`url: string`	scanned, healthy, broken and redirecting links with status	free, account needed

AI visibility

tool	arguments	returns	cost
`ai_visibility_run_scan`	`domain: string`	the prompts asked, per engine presence, and an overall presence score	3,500 credits ($1.75)

Your workspace's AI Visibility data (API key and AI Visibility access required)

tool	arguments	returns	cost
`ai_visibility_get_summary`	`windowDays?: integer (1..365, default 30), engine?: string, region?: string`	visibility, share of voice, engine breakdown, coverage and trend	0 credits per read
`ai_visibility_get_trend`	`windowDays?: integer (1..365, default 30), engine?: string, region?: string`	daily visibility scores and answer counts; missing days are omitted	0 credits per read
`ai_visibility_top_cited_domains`	`windowDays?: integer (1..365, default 30), engine?: string, limit?: integer (1..50, default 20)`	cited domain counts from up to 1,000 recent answers	0 credits per read
`ai_visibility_list_answers`	`windowDays?: integer (1..365, default 30), engine?: string, limit?: integer (1..20, default 20), offset?: integer (0..10000, default 0)`	stored answer verdicts and nextOffset; excludes raw answer text	0 credits per read
`ai_visibility_list_prompts`	`enabledOnly?: boolean (default false), limit?: integer (1..50, default 20), offset?: integer (0..10000, default 0)`	tracked prompts, enabled status and nextOffset; text capped at 2,000 characters	0 credits per read

Google Search Console (API key and a connected Search Console property required)

tool	arguments	returns	cost
`google_search_console_list_sites`	`none`	connected properties, permission level, sync and backfill state, synced date range; connected: false with a connect link when none	0 credits per call
`google_search_console_performance_summary`	`days?: integer (1..480, default 28), site?: string, searchType?: string`	clicks, impressions, CTR and position versus the previous period, plus the daily series	0 credits per call
`google_search_console_breakdown`	`dimension: query \| page \| country \| device, days?, site?, searchType?, sortBy?, sortDir?, limit?: integer (1..100, default 25), offset?, minImpressions?`	clicks, impressions, CTR and position grouped by one dimension, paginated	0 credits per call
`google_search_console_top_queries`	`days?, limit?: integer (1..100, default 25), minImpressions?, site?`	the web queries the site gets clicks for, by clicks	0 credits per call
`google_search_console_top_pages`	`days?, limit?: integer (1..100, default 25), site?`	the site's best performing pages in web search, by clicks	0 credits per call
`google_search_console_striking_distance`	`days?, site?, dimension?: query \| page, minPosition? (default 4), maxPosition? (default 20), minImpressions? (default 50), limit?, offset?`	queries or pages ranking just off the top, sorted by impressions	0 credits per call
`google_search_console_ai_queries`	`days?, site?, searchType?, shape?: conversational \| long_tail \| head \| site_probe, zeroClickOnly?: boolean, minImpressions?, limit?, offset?`	per-shape query totals and classified queries, a sample of assistant-driven demand	0 credits per call
`google_search_console_index_coverage`	`site?: string`	indexed versus not indexed inspected pages, coverage reasons, sitemap versus index diff	0 credits per call
`google_search_console_pages_index_status`	`site?, isIndexed?: boolean, limit?: integer (1..100, default 25), offset?`	inspected pages with Google's verdict, coverage state, canonicals and last crawl	0 credits per call
`google_search_console_list_sitemaps`	`site?: string`	submitted sitemaps with dates, Google's warnings and errors, and url counts	0 credits per call
`google_search_console_inspect_url`	`url: string (inside a connected property), site?: string`	Google's live URL Inspection verdict, canonicals, last crawl, rich results, and the remaining daily budget	0 credits, Editor key, 20 calls per 10 minutes
`google_search_console_refresh_candidates`	`limit?: integer (1..25, default 5)`	published articles losing traffic, each with the reason and the recommended fix	0 credits per call
`google_search_console_sync_status`	`site?: string`	the sync run in flight, recent runs, backfill progress and synced date range	0 credits per call

AI visibility monitors (API key required, no AI Visibility plan; 0 credits per call, each run costs 300 credits per engine x region answer, more on expensive model APIs)

tool	arguments	returns	cost
`monitor_prompts_create`	`prompt: string, cadence: DAILY \| WEEKLY \| BIWEEKLY \| MONTHLY, engines?, regions?, dayOfWeek?, dayOfMonth?, timeOfDay?, timezone?, notifyOn?, useWorkspaceWebhooks?, webhookUrl?, name?, runImmediately?`	the monitor, a monthly cost estimate, and the webhook signing secret once	0 credits per call, Editor key; each run 300 credits per engine x region answer, times the engine's multiplier
`monitor_prompts_list`	`status?: ACTIVE \| PAUSED \| CANCELED \| EXHAUSTED, limit?: integer (1..50, default 20), offset?`	the workspace's monitors with schedule, status, next and last run, credits per run	0 credits per call
`monitor_prompts_get`	`monitorId: string`	one monitor's prompt, engines, regions, schedule, status and notification settings	0 credits per call
`monitor_prompts_update`	`monitorId: string, plus any create field or status?: ACTIVE \| PAUSED`	the updated monitor, with a signing secret when a first webhook is set	0 credits, Editor key
`monitor_prompts_delete`	`monitorId: string`	the canceled monitor; it and its runs stay readable, repeats are safe	0 credits, Editor key
`monitor_prompts_run_now`	`monitorId: string`	one run fired outside the schedule (RUNNING, or SKIPPED when the wallet refused it)	0 credits per call, Editor key; each run 300 credits per engine x region answer, times the engine's multiplier
`monitor_prompts_list_runs`	`monitorId: string, limit?: integer (1..20, default 10), offset?`	runs with per engine x region mentions and citations, change versus the previous run, credits charged	0 credits per call
`monitor_prompts_get_run`	`monitorId: string, runId: string`	one run's per engine x region results and which verdicts changed	0 credits per call

Agents can read this catalog as plain markdown at /mcp/skill.md, the same document your agent is told to read when you paste the install prompt.

// connect

Connect it to your client in under a minute

Add the server URL to your client and sign in to pick a workspace. Headless clients take a workspace secret key instead; keep configurations containing a key out of source control.

Command line (recommended)

Run the command below in your terminal
Run /mcp inside Claude Code, pick omnifamous and choose Authenticate
Sign in to OMNIFAMOUS in the browser, pick a workspace and choose Editor or Viewer access

claude mcp add --transport http --scope user omnifamous https://mcp.omnifamous.com

With an API key

For CI or headless runs, pass a workspace key as a header instead
Run /mcp inside Claude Code to confirm the server is connected

claude mcp add --transport http --scope user omnifamous https://mcp.omnifamous.com \
  --header "Authorization: Bearer sk_your_workspace_key"

--scope user makes the server available in every directory (without it, Claude Code keeps it in local scope and it only appears in the directory where you ran the command). Use --scope project instead to write a .mcp.json for the repo. Without a header that file holds no secret, so it is safe to commit and each teammate signs in.

MCP server vs skill vs plugin vs subagent

The Claude Code ecosystem has four ways to extend an agent, and they are not interchangeable. The short version:

MCP server

A service your agent connects to for data and actions. The client discovers each tool and its arguments through the MCP protocol.

Skill

Instructions that explain a workflow and how to use available tools. Our skill includes the server catalog and connection details.

Plugin

A bundle. One installable package that can contain skills, hooks, subagents and MCP servers together. A distribution format, not a capability.

Subagent

A separate agent assigned part of a task. It can use MCP tools when the host application grants it access.

// security

You are handing an agent access. Here is how we scope it.

Access to one workspace

Each sign-in, or each secret key, covers exactly one workspace. Disconnect an app in Settings, Connected apps, or revoke a key in MCP & API, whenever needed. The agent can use allowed tools but cannot change billing settings.

A free first check

Static page inspection, robots/sitemap checks and cached AI visibility snapshots cost no credits once you connect. Paid tools require Editor access and credits.

Every call itemized

See recent metered calls, reserved credits and completed charges in Billing. Admins can enable automatic refill with a monthly limit on automatic charges.

// pricing, in plain language

Shared credits. Clear prices. Refills you control.

Try a page check

Free with a sign-in

Inspect static page tags and check robots.txt and sitemap files. Rate limited per workspace. No signup credit grant.

Research credits

2,000 credits = $1

Add $5 to $500 to your workspace. Credits roll over. Optional automatic refill has a threshold, an amount and a monthly limit on automatic charges.

Existing visibility reports

0 credits per read

Read your saved monitoring summaries, trends, prompts and answer verdicts. A connected workspace and AI Visibility access are required. Included subscription work is not charged again.

Things to ask once it is connected

// faq

Questions people actually search

What is an SEO MCP server?: An SEO MCP server connects your AI agent to keyword research, traffic estimates, backlink data and AI visibility reports. Connect your workspace once, ask in plain language, and get structured data your agent can turn into a useful analysis.
How do I connect?: Add https://mcp.omnifamous.com to your AI client and sign in when it asks: pick your workspace and Editor or Viewer access. Clients without browser sign-in (n8n, Zapier, scripts, CI) use a workspace secret key from MCP & API settings, sent as an Authorization: Bearer header.
What costs no credits?: Static page inspection, robots/sitemap checks, cached AI visibility snapshots and the engine list cost no credits once you sign in, subject to per workspace rate limits. There is no automatic signup-credit grant.
How do credits work?: One workspace balance is shared by your team and agents. 2,000 credits equal $1 of usage value. Metered tools reserve their quoted price and charge on a successful result; failed calls release the reservation. Purchased and plan-granted credits roll over.
Can credits refill automatically?: A workspace admin can explicitly save a card and enable automatic refill, with a low-balance threshold, refill amount and monthly limit on automatic card charges. It starts off. Low balances are checked every five minutes.
Does this charge again for my subscription?: Scheduled work included in Content Writer and AI Visibility stays covered by those subscriptions. Workspace visibility reads have no per-call credit charge and require AI Visibility access. Buying credits alone does not unlock that subscription.
Can the agent draft or publish articles through this server?: This release exposes research, page checks and existing visibility reports. Article generation and publishing are not in the public MCP catalog.
Can I run an arbitrary AI prompt?: The domain visibility check generates its own prompts. The workspace tools read results already captured by monitoring. Custom prompt execution is not in this release.
Can I give an agent unlimited access?: Use a revocable workspace key and keep it private. Requests are rate limited and metered work is bounded by the workspace's credit balance and spend limits. Agent keys cannot enable automatic payments or change billing settings.

Connect it once. Every session after is smarter.

Connect your workspace