Think inside your AI world.

Datadog through Unl

Datadog shows the monitor is red and the metric that tripped it. Whether that's your incident is a threshold you set, and 'the graph moved' isn't the same as 'you're paged'.

Datadog through Unl reads monitor and metric state against the escalation threshold the platform lead has ratified, so 'is this my incident?' comes back verdict-shaped, naming the condition still unmet.

What Datadog holds

What Datadog through Unl actually reads:

  • Monitor status and the condition that tripped it
  • Metric time series behind the alert
  • APM traces and log lines around the event
  • Dashboards summarising the affected service
  • Security signals flagged in the same window

What the naked read gives you

Ask Datadog through its MCP server what's happening and it answers fully: the monitor that tripped, the metric time series behind it, the APM traces and log lines in the same window, the dashboard for the affected service. That's a genuine, rich read - Datadog's own data, current, unfiltered. What it doesn't do is tell you whether the movement matters to you: the API doesn't know that this service is customer-facing, that the team agreed a five-minute blip isn't worth a page, or that last quarter's postmortem raised the bar for what counts as urgent. The monitor is red. Whether that's your problem is a separate question.

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

What changes when Datadog is read measured

The platform lead ratified in Unl that a monitor only counts as page-worthy when it's customer-facing and the error-budget burn exceeds the agreed rate for more than five minutes - because a single noisy blip was waking the on-call for nothing last quarter.

'Is this my incident?' returns, measured: not yet - the metric moved, but the burn hasn't held past the five-minute mark the team set. Datadog supplied the monitor state and the metric history; your ratified rule supplied the threshold that decides whether it's worth a page.

And back again

The verdict doesn't vanish once it's given. When 'not yet' comes back on a Datadog alert, Unl logs the monitor, the burn rate at the time and the reasoning against the ratified threshold - so next time the same service throws a borderline alert, the on-call sees not just the metric but the precedent, and can ratify a tighter or looser threshold if the pattern's changed.

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 Datadog alert is actually my incident?

Not on its own. Datadog through Unl reads the monitor and metric state; the verdict comes from the escalation threshold you ratified in Unl - customer-facing scope and burn rate held past five minutes. Change the rule and the same alert can return a different verdict.

How do I connect Datadog to Claude?

Through Datadog's own MCP Server, documented at docs.datadoghq.com/mcp_server. Point an MCP-compatible client at it with a scoped API key - the criterion still needs to come from Unl, since Datadog's server only supplies the data.

Does Unl silence or acknowledge the alert?

No. Unl only reads through Datadog's MCP server - monitor state, metrics, the current condition. Silencing, muting or acknowledging an alert would be a write action, and Unl's connection to Datadog doesn't carry one.

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.