Leadstore now has an MCP (Model Context Protocol) connector. Connect it once, in either Claude or ChatGPT, and ask about your site's AI visibility the way you'd ask a coworker. The assistant pulls live data from your account and answers inline, instead of guessing from training data.
What an MCP connector actually does
MCP (Model Context Protocol) is an open standard for letting an AI assistant call out to a real service and pull real data mid-conversation, instead of guessing from training data or asking a user to paste in numbers. Anthropic built it, but it is not Claude-only: OpenAI supports the same protocol for ChatGPT plugins. That is why Leadstore built one connector that works with both, rather than a Claude-only or ChatGPT-only integration.
Connecting is a normal OAuth login, the same kind used to sign into a website with Google. A user approves access once, using their existing Leadstore account, and can revoke it at any time. Nothing is shared across accounts and no token is cached outside that user's own session.
Why this beats asking for its opinion
Ask Claude or ChatGPT about a brand's AI visibility without a connector, and the answer is built from training data and whatever the model happens to know about that industry. It will not know the site's actual visibility score, will not know which tracked competitor is real versus auto-added by mistake, and has no idea what is sitting in the content calendar right now.
Connect the MCP server, and the assistant stops guessing. It asks Leadstore for the number and reports back what is actually true for that account, as of the moment it was asked. A wrong guess dressed up as a confident answer is worse than no answer, and that is the exact failure mode this connector closes.
What you can actually ask it
The connector is read-only today. Seven tools, mapped to seven things a user would otherwise check by logging into the app: visibility score, competitors list, content calendar, blog drafts, brand mentions across AI platforms, share of voice against tracked competitors, and connected sites.
"What's my AI visibility score for my site, and how does that compare to last month?" pulls both numbers in one call, in Claude or ChatGPT.
Set it up
■
In Claude: Settings, Connectors, Add custom connector, paste the server URL. Claude walks through the login.
■
In ChatGPT: Leadstore.xyz is a published plugin. Open Plugins, search for leadstore, press Install plugin and sign in. No Developer Mode and no URL to paste.
■
Both work on free plans, with no plan requirement on either side.
What it will not tell you
The connector deliberately keeps billing, plan name, and subscription status out of anything an assistant can read. Asked what plan an account is on, it points back to the pricing page or account settings instead. That is a design choice, not a technical limitation: anything an MCP tool returns lands in a transcript with a third-party assistant, and subscription details do not belong there. Quota counters, like drafts remaining this month, are the one exception, so the assistant can warn before a request fails rather than after.
Where this goes next
Every tool right now is read-only. The next step is action tools, so a user could tell Claude to kick off an audit or draft a blog post directly, instead of only asking questions. The authentication work to support that safely is already in place, since every read request already runs through a per-user check. What is left is deciding which actions are safe to hand to an assistant in one step, and which need a confirmation prompt first.
How this fits into the three Leadstore stages
■
Stage 1, account + site added: users get the audit reward before any tag or key - four core channel scores, top fixes, readiness, generated llms.txt and llm-data.json, and the Leadstore Tag snippet to install.
■
Stage 2, tag installed: users get the live telemetry reward with no key required - AI-crawler activity by engine, verified-bot counts, human traffic, top sources, top pages, devices and geography.
■
Stage 3, key connected: users add answer tracking on top of the audit and telemetry layers - Visibility Score, SAIV, Brand Monitor, share of voice, citations, competitors and hallucination checks.
Next step
Connect the Leadstore MCP server in Claude or ChatGPT and ask about your AI visibility directly.
