Think inside your AI world.

Neon through Unl

Neon's MCP server will run the query and lay out the schema diff between two branches. Whether that diff is one you would want to ship into main is a call only your migration review makes - drift isn't the same as damage, but it isn't nothing either.

Neon through Unl reads live schema and branch diffs against the migration-parity rule you have ratified, so 'is this branch safe to merge?' comes back ready-or-hold with the condition it hasn't met.

What Neon holds

Point Neon's MCP server at a branch and it returns:

  • SQL query and transaction results
  • table and column schema inspection
  • schema comparisons between branches
  • project, branch, and organisation listings
  • read-only access via a scoped connection URL

What the naked read gives you

Neon's MCP server will run the query, show the transaction result, and lay out the schema diff between two branches field by field. Ask it to compare a feature branch against main and it tells you exactly which tables changed, which columns are new, which types shifted - all without saying whether that diff is one you would want to ship.

The frame judges the data it is given; it does not verify the source’s accuracy.

What changes when Neon is read measured

A founding engineer at a fintech tool ratified a merge rule in Unl: a feature branch only merges to main once its schema diff matches the migration file reviewed in the pull request, because an undocumented column has reached production twice this quarter without anyone noticing until a query broke.

'Is this branch safe to merge into main?' returns, measured: not yet - the branch schema read shows a nullable retry_count column that the migration plan never mentions. Neon supplied the schema diff; your ratified rule supplied the parity bar.

And back again

Fold the column into the migration file, or drop it; Neon through Unl re-reads the branch on its next check, and the ratified parity rule waves it through - the existing page updates, no second one appears.

The answer comes back measured against what you already decided, and why.

The lane is live and open to this tool today: 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.

Questions people ask

What would move this from 'not yet' to 'cleared'?

Making the branch schema match the reviewed migration - either by amending the migration file or removing the untracked column. The same ratified parity rule reads the reconciled diff and clears the branch.

How do I connect Neon to Claude?

Add Neon's managed MCP server via OAuth in Claude's MCP settings, or run the local server with your own Neon API key instead - no separate installation needed either way. Either route, Unl's reads pass through that same server.

Does Unl write to my database?

No. Unl reads query results and schema state through Neon's MCP tools; you can also scope the connection to read-only. Any migration or branch merge, you run yourself.

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.