Repository avatar
AI Tools
v1.1.0
active

promptmentions

com.promptmentions/promptmentions

Read-only AI visibility data: brand mentions, cited sources and answers across AI engines.

Documentation

PromptMentions MCP server

Connect an AI assistant to your PromptMentions AI-visibility data.

PromptMentions tracks what ChatGPT, Claude, Gemini, Grok, Meta AI, Perplexity and Google's AI Overviews and AI Mode say about a brand. It asks your buyers' questions on a schedule and keeps every answer with the pages it cited. This MCP server gives an assistant read-only access to one workspace's results.

This repository holds connection instructions and directory metadata only. The server is hosted by PromptMentions; there is nothing to install or run.

Connect

Server URLhttps://promptmentions.com/api/mcp
TransportStreamable HTTP
Sign-inOAuth 2.1, with dynamic client registration and PKCE. An API key also works.
AccessRead-only: it cannot change anything, run checks or spend your allowance

With OAuth (recommended)

Add the server URL to any MCP client that supports remote servers, such as Claude, Claude Code, Cursor or VS Code. The client opens a PromptMentions sign-in page. You choose the workspace, and the clients it may read, then press Allow. To disconnect the app, go to Settings › API & MCP › Connected apps.

Claude Code:

claude mcp add --transport http promptmentions https://promptmentions.com/api/mcp

With an API key

Create a key in Settings › API & MCP, then send it as a header:

{
  "mcpServers": {
    "promptmentions": {
      "type": "http",
      "url": "https://promptmentions.com/api/mcp",
      "headers": { "Authorization": "Bearer YOUR_API_KEY" }
    }
  }
}

Without a key or a sign-in, only initialize, ping and tools/list answer, so a client can see what the server offers before you connect.

Tools

All tools are read-only.

ToolWhat it returns
get_workspaceThe workspace, the connection's permissions and the client projects it can read
list_projectsClient projects with their brand, aliases, tracked competitors and locale
get_projectOne client project
list_prompt_setsPrompt sets for a client, with every saved revision, its locale and prompts
list_runsMonitoring runs for a client, newest first, with sample counts
get_run_metricsOne run's labelled metrics (mention rate, citation rates, mention share) and competitor share of voice
list_samplesEvery answer in a run: provider, model, prompt, locale, time, status, mentions and citations
get_citationsDomains and pages cited in a run's answers, classified, with citation gaps
get_sample_evidenceThe saved evidence of one answer: answer text, citations and extraction summary
list_opportunitiesTasks and opportunities for a client, with their source evidence
get_looker_rowsFlat rows for Looker Studio and spreadsheets
get_reportA client report for a date window: runs with labelled metrics, opportunity changes and caveats

Each figure is labelled with the AI engine, the date and the number of answers behind it. Failed answers are left out.

Links

No installation packages available.