Docs/AI Visibility/Agency MCP

AI Visibility MCPCloud

Run S6S AI Visibility from inside your own AI agent. Audit a prospect, register and check a brand across ChatGPT, Gemini, and Perplexity, pull share of voice and fix plans, and assemble a client-ready proposal — all from Claude, Cursor, or any MCP client.

Built for agencies

@s6s/visibility-mcp is a read-only bridge to the S6S AI Visibility API. Every tool is an authenticated call that returns the same confidence-labeled data you see in the Report — so the model you already work in can do the audit, the comparison, and the proposal without you leaving the chat.

It is measurement and planning only: there are no site-modifying tools. Fix plans are drafts your client approves and applies — S6S never publishes anything automatically.

Quick start

1. Get a free API key

Create one at s6s.ai → Settings → API Keys. The free readiness_snapshot works with no key at all — every other tool needs one.

2. Add it to your AI client

Add S6S to your MCP config — claude_desktop_config.json, .cursor/mcp.json, or your client’s equivalent:

claude_desktop_config.json / .cursor/mcp.json

{
  "mcpServers": {
    "s6s-visibility": {
      "command": "npx",
      "args": ["-y", "@s6s/visibility-mcp"],
      "env": { "S6S_API_KEY": "sk_your_key_here" }
    }
  }
}

3. Try it without any setup

Run the free readiness snapshot straight from the terminal — no key, no install:

npx -y @s6s/visibility-mcp

Tools

readiness_snapshotFree · no key

Deterministic AI-readiness audit on any domain — crawler access (GPTBot, Google-Extended, PerplexityBot), schema, llms.txt, Wikipedia/Reddit presence. The free prospect hook; zero credits.

register_brandFree

Auto-discover and register a brand from its domain (name, category, keywords, competitors). Does not start monitoring or spend credits. Returns the brandId the other tools use.

check_visibilityCredits

Run a full multi-engine check across ChatGPT, Gemini, and Perplexity — AI App, Local, and Search & Agent surfaces. Polls until complete.

get_reportFree

The latest per-surface report, with confidence states and which engines mention the brand.

get_share_of_voiceFree

Competitive presence — who AI recommends instead — carrying the report's directional/measured labels.

get_fix_planFree

Prioritized, engine-specific recommendations tied to the report evidence. Drafts the client approves; never auto-published.

get_attributionFree

Before/after impact for fixes already applied, with the confidence floor — for renewals and retention.

assemble_proposalPrompt

A guided prompt that pulls the report + share of voice + fix plan and writes a confidence-honest, client-ready proposal.

The agency flow

Snapshot the prospect

Run readiness_snapshot on their domain — a free, instant read of how prepared their site is for AI engines. Use it as the hook in the first conversation.

Register & check

register_brand auto-discovers them, then check_visibility runs the full multi-engine check — what ChatGPT, Gemini, and Perplexity actually say about them for their real buyer queries.

Show the gap

get_report and get_share_of_voice give you the per-surface scores and who AI recommends instead — the 'here's where you stand' story, with confidence labels intact.

Propose & prove

get_fix_plan is your plan; assemble_proposal turns it all into a client-ready proposal. After work ships, get_attribution shows what moved — the renewal story.

How to present the data

Quote S6S’s scores and confidence labels exactly — an “early read” or “directional” result is never a confident number.
Recommendations are expected to improve readiness, not a guaranteed visibility lift. Outcomes depend on factors no one controls.
Fixes are applied with human approval. Nothing is published autonomously.
Only high-confidence movements may be claimed as attributed impact; medium is “likely contributed”.

AI Visibility docs Inside the Report Create an API key