Labsco
open-strategy-partners logo

CitedSpy

from open-strategy-partners

Connect AI tools like Claude, Cursor, and Codex directly to your CitedSpy workspace with the Model Context Protocol (MCP). Once connected, your AI assistant can read your brands, prompts, runs, citations, and analytics - and create prompts or trigger runs - without you copying data back and forth. You can ask for your AI visibility dashboard in plain language, find your weakest prompts and engines, create and edit tracking prompts from chat, trigger fresh runs across engines, pull citations, mentions, and reports into your AI workflow

πŸ”₯πŸ”₯πŸ”₯βœ“ VerifiedFreeQuick setup

Integrations

Connecting AI tools with MCP

Connect AI tools like Claude, Cursor, and Codex directly to your CitedSpy workspace with the Model Context Protocol (MCP). Once connected, your AI assistant can read your brands, prompts, runs, citations, and analytics - and create prompts or trigger runs - without you copying data back and forth.

What you can do

  • Ask for your AI visibility dashboard in plain language

  • Find your weakest prompts and engines

  • Create and edit tracking prompts from chat

  • Trigger fresh runs across engines

  • Pull citations, mentions, and reports into your AI workflow

Step 1 - Create an API key

  • Log in to app.citedspy.com.

  • Go to Settings > API Keys.

  • Click New API Key, name it (for example "Claude MCP"), and copy the key.

Your key starts with cspy_ and is only shown once. API keys are available on the Growth and Pro plans.

Step 2 - Connect your AI tool

Pick your tool below. Every tool points at the same server URL (https://mcp.citedspy.com/mcp) and finishes with the same one-time authorization in Step 3.

Claude Code (CLI)

Run this in your terminal:

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

Then start Claude Code, run /mcp, choose citedspy, and authorize.

Claude Desktop

  • Open Settings > Connectors > Add custom connector.

  • Name it CitedSpy and paste the URL https://mcp.citedspy.com/mcp.

  • Click Add, then Connect to authorize.

Gemini CLI

Run:

gemini mcp add --transport http citedspy https://mcp.citedspy.com/mcp

Restart Gemini CLI, run /mcp, and follow the prompt to authorize citedspy. If your version has no mcp add command, add the server to ~/.gemini/settings.json instead:

{
 "mcpServers": {
 "citedspy": {
 "httpUrl": "https://mcp.citedspy.com/mcp"
 }
 }
}

Codex (CLI and Desktop)

Add this to ~/.codex/config.toml:

[features]
experimental_use_rmcp_client = true

[mcp_servers.citedspy]
url = "https://mcp.citedspy.com/mcp"

Then authorize from the terminal:

codex mcp login citedspy

The rmcp_client feature is required for remote servers that use sign-in (OAuth), like CitedSpy.

Cursor

Add the server to ~/.cursor/mcp.json (global) or .cursor/mcp.json (project):

{
 "mcpServers": {
 "citedspy": {
 "url": "https://mcp.citedspy.com/mcp"
 }
 }
}

Cursor opens your browser to authorize the first time the server is used.

Any other MCP client

Any tool that supports remote (streamable HTTP) MCP servers works. Add a new server with the URL https://mcp.citedspy.com/mcp and complete the authorization in Step 3.

Step 3 - Authorize CitedSpy

The first time you connect, your AI tool opens a CitedSpy authorization page in your browser. Paste your API key, confirm the connection, and return to your AI tool. CitedSpy validates the key before the connection is created, and never exposes it to the AI tool. You only do this once per tool.

Example prompts

List all the brands in my CitedSpy workspace.
Show the visibility dashboard for my main brand over the last 30 days.
Create a comparison prompt for my brand against its top two competitors, then trigger a run.

Available actions

Action What it does ping Validates your key and shows your workspace and plan list_brands / get_brand Read brands and their competitors list_prompts / get_prompt / get_prompt_report Read prompts and full analytics reports create_prompt / update_prompt / delete_prompt Manage tracking prompts list_runs / get_run / trigger_run Read runs and queue new ones list_competitors Read competitors for a brand get_dashboard Read computed dashboard metrics list_citations / list_mentions Read citations and detected mentions list_engines Read engines and your on/off selection call_citedspy_v1 Call any API endpoint directly

Security

Your API key authorizes MCP access - treat it like a password. The MCP can create, update, and delete prompts and trigger runs, so only connect AI tools you trust. To revoke access, disconnect the server in your AI tool or delete the key in Settings > API Keys.

Need more help?

Our support team is here to help you get the most out of CitedSpy.

Contact support