Sleepwalker MCP server
{
"mcpServers": {
"sleepwalker": {
"type": "http",
"url": "https://mcp.sleepwalker.ai/mcp"
}
}
}{
"mcpServers": {
"sleepwalker": {
"type": "http",
"url": "https://mcp.sleepwalker.ai/mcp"
}
}
}{
"mcpServers": {
"sleepwalker": {
"type": "http",
"url": "https://mcp.sleepwalker.ai/mcp"
}
}
}{
"servers": {
"sleepwalker": {
"type": "http",
"url": "https://mcp.sleepwalker.ai/mcp"
}
}
}{
"mcpServers": {
"sleepwalker": {
"type": "http",
"url": "https://mcp.sleepwalker.ai/mcp"
}
}
}file .mcp.json (in your project root)file claude_desktop_config.jsonfile ~/.cursor/mcp.jsonfile .vscode/mcp.jsonfile ~/.codeium/windsurf/mcp_config.jsonversion 0.1.1
What it does
Sleepwalker provides tools for AI visibility and content intelligence. It exposes capabilities to run prompts across AI search platforms (ChatGPT, Perplexity, Grok, Gemini), capture citations and mentions, export page content as structured markdown, and audit how AI crawlers see your pages.
- Monitor how AI search platforms mention your brand across different prompts and platforms.
- Export public web pages as markdown bundles with both human-readable content and technical crawlability data for use in AI agent workflows.
- Audit pages to discover what structured data, metadata, and content an AI crawler actually receives versus what readers see.
Tools this server exposes
| Tool | Does |
|---|---|
| visibility_run | Run a prompt across specified AI platforms (OpenAI, Perplexity, Grok, Gemini) and capture responses, citations, and competitor mentions. |
| okf_export | Export a public page to Open Knowledge Format markdown files with content and technical snapshots, runnable without authentication. |
| content_intelligence | Serialize public pages, score content depth and freshness, and return recommendations for improvement. |
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.