Your agents don’t need to persist. Your purpose does.
Can AI grade my OKRs against my own rule?
A model can compute completion and hand you a number between nought and one. That is a calculation, not a grade. Grading needs your rule — what met, partial and missed mean, and why — and only then does the score reflect your standard instead of a generic ratio. Through Unl the rule is applied.
Unl, your why agent. The work behind “Can AI grade my OKRs against my own rule?” depends on decisions your AI needs to inherit. You settle what matters and why; Unl brings the relevant decisions and why into the next session, across the AIs you use. Your agents can then act on your decisions, check the result and carry the outcome forward. They return to you when intent is unresolved, decisions conflict or the action needs your identity. Use Unl with your AI or see how Unl works.
A ratio is not a grade
Completion over target gives a tidy number, but it flattens judgement you actually hold: that some key results deserve partial credit only past a real threshold, that others are all-or-nothing, that a near-miss on a critical goal is worse than a near-miss on a stretch one. A flat ratio ignores all of it.
So the generic score is precise and shallow. It looks objective and quietly encodes a scoring philosophy — linear, uniform — that is probably not yours. Grading against your rule is what replaces a borrowed philosophy with your own.
Grading against your standard
Say your scoring rule is ratified: committed key results grade as met only at ninety per cent and above, stretch key results earn partial credit from fifty, and a missed committed goal counts double against the objective, because commitments are promises and stretches are bets. The AI applies that rule to each result and returns grades that reflect it.
So a stretch key result at sixty per cent grades as a genuine partial, and a committed one at eighty-five grades as a miss despite the high number — because your rule says so. The grades are yours, defensible by the standard you set, not a uniform ratio.
Grades that hold up
Because the rule was ratified before the results, the grades cannot be argued into something else after the fact — they follow the standard. And because the reason travels with the rule, each grade explains itself: this is a miss because it was a committed goal below the met line, which you weighted for a reason.
So “can AI grade my OKRs against my own rule?” is yes in the only sense worth having: the grade is produced by your standard, applied consistently, with the reasoning attached — not a generic number you then have to defend.
AI can compute a completion ratio, but grading needs your rule — what met, partial and missed mean, and why; through Unl the AI applies the scoring rule you ratified before the results, so the grades reflect your standard, explain themselves, and hold up — not a generic nought-to-one that encodes a philosophy that isn’t yours.
Reads through Unl arrive with measured context — in the presence of the decisions you’ve already settled. It works 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.
Read further
Questions people ask
Can AI grade my OKRs against my own scoring rule?
Yes, once the rule is present. A model can compute completion over target, but that is a calculation that treats every goal the same. Grading needs your rule — what met, partial and missed mean, and why — and through Unl the AI applies that ratified rule, so the grades reflect your standard.
Why isn't a completion percentage a real OKR grade?
Because it flattens judgement you hold — that some key results deserve partial credit only past a threshold, that others are all-or-nothing, that a missed commitment is worse than a missed stretch. A flat ratio encodes a uniform philosophy that is probably not yours.
How do I get consistent OKR grades?
Ratify your scoring rule before the results are known and apply it to each key result. Through Unl the grade follows the standard you set rather than the debate, and because the rule predates the number it can’t be argued into something else after the fact.
What does an AI need from me before it can grade an OKR?
Your scoring rule — what met, partial and missed mean, and why. Held in Unl, the rule is applied to each result the same way every quarter, so the grade is your standard computed consistently, not the model’s take on the number.
What this is
Your agents don’t need to persist. Your purpose does.
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 works inside Claude, Claude Code, ChatGPT and Cursor, and connects 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 to start. Paid plans from $10 a month.