Think inside your AI world.
The project health review, through Unl
A health review gathers people to agree on a colour. Through Unl the read applies the conditions you actually set — what makes it green, amber, red — to the state as it is now, and returns the verdict those conditions imply. The colour stops being chosen and starts being computed against your own rules.
The health review’s output is a colour, and a colour is only honest if the conditions behind it were applied. Unl holds those conditions, so the review opens with the verdict your rules produce — not a swatch someone picked, but green-or-not measured against your thresholds, current as of the read. Here the verdict replaces the chart.
What is the health review deciding?
Nominally, a colour; actually, a judgement the colour stands for. Is the project where your conditions say it should be? That judgement needs the conditions applied to current state — and in most reviews the conditions live in someone’s memory and the state lives in the tools, so the colour gets negotiated rather than computed.
A negotiated colour is a mood with a swatch. It can be talked up before a stakeholder or down before a planning cut, because nothing forces it to match the conditions you set.
What does a measured health review return?
Say your conditions are fixed: green needs no severity-one incident open and week-one retention at or above 35%. The review opens with the read applying both to the current state: “Not green — a severity-one is open, which fails the first condition; retention is 31% against your 35% line.” The colour is computed, not chosen.
A model can restate last week’s colour but can’t compute this one: your two conditions are a decision, not a metric it tracks. Measured context supplies them, so the review starts from a verdict everyone can check rather than a colour they have to trust.
What does the meeting become?
An audit, not a negotiation. The verdict is on the table from the first second; the review is where you confirm it, dig into the failing condition, and decide what to do — not where you argue about what colour to write down.
That’s the health review thinned to its purpose: the conditions you set, applied to the current state, so the meeting spends its time on the one condition that failed instead of on the colour itself.
A health review picks a colour that’s only honest if your conditions were applied; through Unl the read computes it against the thresholds you set, current as of now — the verdict replaces the chart.
Reads through Unl arrive with measured context — in the presence of the decisions you’ve already settled. The reach lane is live: 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.
Read further
Questions people ask
How do I run a project health review that isn’t just a colour?
Compute the colour from the conditions you set instead of negotiating it. “Green needs no severity-one incident open and week-one retention over 35%” is a rule that can be applied to the current state; measured context holds the rule and returns the verdict, so the review opens with a colour everyone can check rather than one they have to trust.
Why is our RAG status always up for debate?
Because a negotiated colour is a mood with a swatch — nothing forces it to match your conditions, so it can be talked up before a stakeholder or down before a cut. When the conditions are applied to current state at read time, the colour is computed rather than chosen, and the debate moves to the failing condition.
Can AI keep our health status current?
It can restate last week’s colour, but computing this week’s needs your conditions for green, which are a decision rather than a metric. Measured context supplies them so the read returns a live, checkable verdict. The reach lane is live: 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.
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.