Think inside your AI world.
Timescale through Unl
Timescale's MCP server will read back the age of the oldest row sitting in a hypertable. Whether that age breaches the retention policy you're supposed to be running is a ceiling your own rule holds - a query result isn't the same as a policy check.
Timescale through Unl reads a hypertable's oldest live chunk timestamp against the retention rule you have ratified, so 'does this table actually satisfy the retention policy?' comes back flagged with the condition it hasn't met.
What Timescale holds
Ask Timescale's Tiger CLI MCP server about a database service and it returns:
- SQL query results run against a database service
- the full database schema, including hypertables and continuous aggregates
- the list of database services in a project
- detailed configuration for a specific service
- log output for a running service
What the naked read gives you
Tiger CLI's MCP server for Timescale runs a SQL query against a database service on request, and lays out that service's schema - tables, hypertables, continuous aggregates, indexes, and how they're built - as readable text. It can tell you exactly how old the oldest row in a hypertable is and exactly how the schema is configured, without saying whether that configuration is the one you're supposed to be running.
The frame judges the data it is given; it does not verify the source’s accuracy.
What changes when Timescale is read measured
A data engineer at an industrial IoT monitoring company ratified a retention rule in Unl: no hypertable may hold raw, uncompressed sensor readings older than ninety days in the primary schema, because a routine audit once found three-year-old raw telemetry still sitting there, inflating both storage cost and audit exposure.
'Does this table actually satisfy the retention policy?' returns, measured: not yet - the schema and query read shows the oldest row in sensor_readings dates back four hundred and twelve days, well past the ninety-day ceiling the rule sets. Timescale supplied the chunk age; your ratified rule supplied the ceiling.
And back again
Running the retention policy - or dropping the aged chunks by hand - is what closes the gap. The next schema read finds nothing older than the ceiling, and Timescale through Unl marks this same page clear against the rule.
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 table against the retention rule?
Dropping or compressing chunks older than ninety days, typically by running the configured retention policy. The same ratified rule reads the newer oldest-row timestamp and clears the table.
How do I connect Timescale to Claude?
Install Timescale's official Tiger CLI (timescale/tiger-cli), authenticate against your Tiger Cloud project with tiger auth login, and run its built-in MCP server, then register it in Claude's MCP configuration - Unl's reads travel through that same server.
Does Unl write to my database?
No. Unl calls db_schema and db_execute_query in read-only mode to see current state; it drops no chunks and changes no retention policy. Any cleanup, 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.