Skip to main content
Mentionpath exposes a Model Context Protocol (MCP) server so AI assistants can read your AI visibility data and propose changes (new prompts, audits, Webflow drafts) with the same permissions and approval rules as the REST API.

​
Endpoint

URLhttps://api.mentionpath.ai/mcp
TransportStreamable HTTP (POST only)
AuthOAuth 2.1 with PKCE (hosted clients) or bearer token
Call workspace_how_to_use after connecting for a short playbook, or workspace_list_domains to list domain IDs for domain-scoped tools.

​
What you can do

Read analytics

Visibility KPIs, platform comparisons, brand rankings, mention and citation analytics, and traffic summaries.

Manage prompts & audits

List prompt definitions, trigger site audits, and review open issues (writes require approval by default).

Webflow drafts

Read CMS collections and create or update draft items when Webflow is connected for the domain.

Workflow prompts

Server-authored prompts such as visibility reports and audit prioritization plans.

​
Where to connect in the app

In Mentionpath, open Account → Integrations for MCP client setup guides (Claude, ChatGPT, Cursor, Claude Code, VS Code, Codex, Lovable). API tokens and usage live under Account → API and Account → Usage.
The REST API at /api/v1 mirrors every MCP tool. Use REST for server-side integrations; use MCP when an AI client should call Mentionpath directly.