Think inside your AI world.
Grafana through Unl
Grafana shows the panel and the query behind it. Whether the trend is a regression or normal drift is your call - and a red line isn't the same as a ticket.
Grafana through Unl checks dashboard and datasource queries against the customer-impact definition the SRE has ratified, so 'is this trend a regression?' returns verdict-shaped, naming what's missing.
What Grafana holds
The real read surface behind Grafana through Unl:
- Dashboard panels and the queries behind them
- Datasource queries across Prometheus, Loki and the rest
- Alert rules and their current routing state
- Grafana Incident records tied to the panel
- Sift investigations for error patterns and slow requests
What the naked read gives you
Grafana through its MCP server hands back exactly what you'd see on screen: the panel, the query behind it, the datasource it hit, the alert rule and its current routing. That's real, current data straight from the instance. What it doesn't supply is the context that turns a wobbling line into a decision - whether this particular service's dashboard is one the team watches for customer impact, or one that's always noisy on a Monday. The panel is red. Whether it's a regression is a judgement the dashboard was never built to make.
The frame judges the data it is given; it does not verify the source’s accuracy.
What changes when Grafana is read measured
The SRE ratified in Unl that a dashboard trend only counts as a regression when it crosses the customer-impact line drawn for that specific service - because chasing every Monday-morning spike was burning a day a week on nothing.
'Is this trend a regression?' returns, measured: no - the panel moved, but it sits inside the range the team already classed as normal drift for this service. Grafana supplied the panel and the query behind it; your ratified rule supplied the line that separates drift from regression.
And back again
Each answer writes back into Unl as a small ruling: the dashboard, the panel, the reading against the customer-impact line, and the verdict itself. Ask the same question about the same service next Monday and the SRE isn't starting from a blank panel - the prior read and the line it was measured against are both sitting in Unl, ready to be tightened, loosened, or reused.
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
Can AI tell me if a Grafana trend is a real regression?
Not by itself. Grafana through Unl reads the panel and the query behind it; whether that reading counts as a regression depends on the customer-impact line you ratified in Unl. The same panel returns a different verdict if the line moves.
How do I connect Grafana to Claude?
Grafana publishes its own MCP server at grafana.com/docs/grafana/latest/developer-resources/mcp. Configure it against your instance with the right access, and pair it with Unl - the panel arrives from Grafana, the line it's measured against arrives from your ratified rule.
Does Unl edit the dashboard or the alert rule?
No. The connection to Grafana is read-only in practice: panels, queries and alert state come back, nothing gets written to the instance. Any change to a panel or a routing rule happens in Grafana itself, not through Unl.
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.