Inite Visibility MCP server
{
"mcpServers": {
"inite-visibility": {
"type": "http",
"url": "https://inite.ai/api/mcp"
}
}
}{
"mcpServers": {
"inite-visibility": {
"type": "http",
"url": "https://inite.ai/api/mcp"
}
}
}{
"mcpServers": {
"inite-visibility": {
"type": "http",
"url": "https://inite.ai/api/mcp"
}
}
}{
"servers": {
"inite-visibility": {
"type": "http",
"url": "https://inite.ai/api/mcp"
}
}
}{
"mcpServers": {
"inite-visibility": {
"type": "http",
"url": "https://inite.ai/api/mcp"
}
}
}file .mcp.json (in your project root)file claude_desktop_config.jsonfile ~/.cursor/mcp.jsonfile .vscode/mcp.jsonfile ~/.codeium/windsurf/mcp_config.jsoninstall npx -y @inite/visibilityversion 1.1.2
What it does
This server audits website visibility to AI answer engines (ChatGPT, Claude, Gemini, Perplexity) and returns a score out of 100. It checks crawler access policies, identity files, page signals, and queries the engines directly to see if they can find and name a site. It is for developers and site owners who need to understand how discoverable their websites are to AI systems.
- Check which AI crawlers can access a website by examining robots.txt and identifying which are blocked for retrieval versus training data collection.
- Audit whether a site has published identity files like llms.txt and ai.json that help AI systems identify it.
- Get a comprehensive visibility score by running a full audit that queries four answer engines directly and weights multiple factors into a single number.
Tools this server exposes
| Tool | Does |
|---|---|
| check_ai_access | Returns which answer-engine crawlers are allowed by robots.txt, separated by whether they fetch live pages or only collect training data. |
| check_identity_files | Reports which of the ten identity files exist at a URL, including llms.txt, ai.json, and identity.json. |
| check_page_signals | Extracts page title, description, canonical URL, hreflang tags, and Schema.org JSON-LD types from a website. |
| analyze_site | Starts an asynchronous full audit of a site's visibility to AI engines and returns a run ID; requires authentication and consumes account allowance. |
| get_analysis | Retrieves the progress and results of a running or completed audit, including the final score out of 100 and report URL. |
Before you enable it
An MCP server runs with whatever access you give it. Start with a development account rather than production, and check which tools the server exposes before you let an agent call them unattended.