Think inside your AI world.
CockroachDB through Unl
CockroachDB's managed MCP server will tell you precisely how many queries are hammering a table right now. Whether that's a safe moment to run a migration against it is a window your own rule defines - a query count isn't a green light.
CockroachDB through Unl reads live running-query counts against the migration-safety window you have ratified, so 'is it safe to run this migration now?' comes back provisional with the condition it hasn't met.
What CockroachDB holds
Ask CockroachDB's managed MCP server about a cluster and it reports:
- cluster listings and detailed cluster information
- database and table listings within a cluster
- full schema detail for a given table
- results from SELECT and EXPLAIN statements
- currently executing queries against the cluster
What the naked read gives you
CockroachDB's managed MCP server lists every accessible cluster, reports cluster health, lists the databases and tables inside one, hands back a table's full schema, runs the SELECT or EXPLAIN you ask for, and shows exactly which queries are running against the cluster right now. It gives an accurate, live picture of what the cluster is doing, without saying whether right now is a safe moment to change it.
The frame judges the data it is given; it does not verify the source’s accuracy.
What changes when CockroachDB is read measured
A site reliability engineer at a multi-region payments platform ratified a migration-safety rule in Unl: no schema migration runs against the orders table while more than fifteen queries are actively executing against it, because a migration launched during a traffic peak once locked the table for six minutes and queued a backlog of failed payments.
'Is it safe to run this migration now?' returns, measured: not yet - the running-queries read shows thirty-one active statements against the orders table, more than double the fifteen-query ceiling the rule allows. CockroachDB supplied the running-query count; your ratified rule supplied the safety window.
And back again
There's no code fix here, only timing: wait for the query count to fall, or schedule the migration for a quieter window. Either way, CockroachDB through Unl sees the lower count on its next look, and the page updates in place.
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 clear this migration to run?
The active query count against orders dropping under fifteen, whether that means waiting for off-peak hours or waiting out a traffic spike. The same ratified rule reads the lower count and clears the window.
How do I connect CockroachDB to Claude?
Enable the managed MCP server from your CockroachDB Cloud cluster settings and authorise it via OAuth or a service account API key in Claude's MCP configuration - Unl's reads travel through that same managed endpoint.
Does Unl write to my cluster?
No. Unl calls CockroachDB's list_databases, get_table_schema, select_query, and show_running_queries tools, all introspective by default; the write-enabled tools stay off unless you turn them on separately. Any migration, you run it yourself.
What this is
Think inside your AI world — you stay in command
Save the thoughts, decisions and targets worth keeping, each with its reasoning, carried into every AI session the moment they matter. A new unit of exchange between you and your AI: the Settled Why with standing that travels. Unprompted.
Your whole AI world. What you decided at the epicentre. It plugs into Claude, Claude Code, ChatGPT and Cursor as an MCP connector — quick to connect, in a couple of steps.
MCP native·Human settled·Model agnostic·Your data
Measured Context
Connect Unl to bring the right information into the moment.
Your sources, read against the criteria you set.
The full product, open. Free at launch.