Think inside your AI world.
How do I connect Vercel to Claude?
Vercel's MCP server covers docs, projects, deployments, logs and agent runs — the observational surface of the deploy pipeline, in conversation.
Direct answer: add Vercel's MCP server in Claude via Settings → Connectors and authorise ('claude mcp add' in Claude Code). Claude reads your docs, projects, deployments, logs and agent runs. Deploy questions — what shipped, what failed, what the build log actually says — get answered from the platform rather than from a dashboard detour.
Deploys, read not refreshed
Did the preview build pass, what error killed the production deploy, which project changed today, what the function logs show around the spike — pulled live mid-conversation. The log read during an incident is where this connector earns its place.
The window rule the platform ships through
Worked case: you ratified a deploy-window rule — no production deploys after Thursday noon, because weekend rollbacks are a one-person company's nightmare and you are the one person. Vercel deploys whenever asked; calendars of courage are policy, and the platform holds none.
The pipeline, read inside the window
Through Unl a Friday deploy conversation arrives pre-framed: production push proposed, and your window rule — Thursday noon, the weekend-rollback logic — stands over it. The read stays one-way; Unl deploys nothing. It just makes sure the rule you wrote for tired-Friday-you is present when tired-Friday-you starts typing.
Vercel to Claude reads the deploy surface; Vercel through Unl reads it with your deployment policies standing — the Thursday rule shows up on Friday, which is the whole point of having written it down.
Reads through Unl arrive with measured context — in the presence of the decisions you’ve already settled. The reach lane is live: one box, paste anything. If it speaks MCP, Unl can reach it. Readings arrive unprompted, the data beside the criterion; Unl is a courier, not a warehouse, and keeps only your keys and the frame.
Read further
Questions people ask
What does the Vercel MCP server read?
Docs, projects, deployments, logs and agent runs — the observational layer of the platform. Enough for status, failure diagnosis and what-changed questions.
Does this work from ChatGPT?
Yes; the server speaks standard MCP and any capable client connects the same way. The measured layer through Unl is likewise surface-agnostic, tuned first for Claude and Claude Code.
Can Unl trigger a deployment?
Not while measuring — a read only reads, so nothing Unl does in the course of checking a deploy against your rules touches the platform. But the honest answer is not a flat no: where a server exposes a tool that acts, Unl can call it, because the socket exposes a server's full surface rather than a chosen subset. That call takes your explicit in-turn gesture and is verified at the choke point, and a tool that does not declare itself read-only is treated as a write and gated by default. So a deployment never happens as a side effect of a read; it happens when you ask for it and confirm it.
Does Unl change or store my Vercel data?
Not stored: what Unl reads is quarantined for the turn and never kept — no query text and no read content reaches our telemetry. Not changed by a read, either — a read only reads. Writes exist and are gated: every one carries your explicit in-turn gesture, verified at the choke point, and a tool that does not declare itself read-only is treated as a write by default rather than assumed safe.
What this is
Think inside your AI world — you stay in command
Unlimitless (Unl to friends) holds what you've settled, reads what your tools are showing, and catches what's changed out in the world — and hands your AI whatever bears on the work, the moment it's needed, without you asking. The right thing, in front of the model, unprompted, with you in command of the call. So you keep moving toward what you set out to build, on top of everything you've already decided.
It plugs into Claude, Claude Code, ChatGPT and Cursor as an MCP connector. Quick to connect, in a couple of steps.
Unlimitless is open now to invited Alpha. Apply for the Beta waitlist to come in ahead of the full launch:
Alpha is invite-only · free at launch.