Refd MCP server
{
"mcpServers": {
"refd": {
"type": "http",
"url": "https://api.refd.ai/mcp"
}
}
}{
"mcpServers": {
"refd": {
"type": "http",
"url": "https://api.refd.ai/mcp"
}
}
}{
"mcpServers": {
"refd": {
"type": "http",
"url": "https://api.refd.ai/mcp"
}
}
}{
"servers": {
"refd": {
"type": "http",
"url": "https://api.refd.ai/mcp"
}
}
}{
"mcpServers": {
"refd": {
"type": "http",
"url": "https://api.refd.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 0.2.0
What it does
Refd is a read-only MCP server that exposes AI search monitoring data for brands, including visibility scores, competitor mentions, citations, and answer evidence across AI platforms like ChatGPT, Perplexity, and Gemini. It is aimed at developers and AI agents that need programmatic access to brand-tracking workspace data collected by the refd.ai platform.
- Query how often a brand is mentioned or cited in AI-generated answers across multiple AI search engines.
- Retrieve competitor tracking data and compare mention frequency within a monitored workspace.
- Access prompt-level answer evidence and citation details to audit how AI platforms represent a brand.
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.