Agent Loop Foundry.

Prioritization instrument · from Field Report 01

The Task Scorecard.

Your org runs hundreds of tasks. This ranks any one of them — how safe and easy the win is, whether it’s worth anyone’s time, and whether the investment will last — so you know where to start, what to leave alone, and what’s too risky to touch yet. Rate one real task across all 32 dimensions; it scores the three, applies the caps, checks the five gates, and ranks the task for you, live.

Autosaves to this browser session only — nothing leaves your machine.

Saturation Readiness

Is this task safe and predictable enough to be an easy win — or still too risky and judgment-heavy to hand to AI?

How repetitive it is

weight ×8

Hover or pick 0–5 to preview what each rating means.

Predictable inputs and outputs

weight ×7

Hover or pick 0–5 to preview what each rating means.

How stable the rules are

weight ×6

Hover or pick 0–5 to preview what each rating means.

Easy to tell if it's right

weight ×12

Hover or pick 0–5 to preview what each rating means.

Known-correct answers exist

weight ×8

Hover or pick 0–5 to preview what each rating means.

You find out fast if it worked

weight ×4

Hover or pick 0–5 to preview what each rating means.

Mistakes are easy to fix

weight ×10

Hover or pick 0–5 to preview what each rating means.

A wrong answer isn't costly

weight ×11

Hover or pick 0–5 to preview what each rating means.

A mistake stays contained

weight ×6

Hover or pick 0–5 to preview what each rating means.

Easy for a person to double-check

weight ×6

Hover or pick 0–5 to preview what each rating means.

Better quality stops mattering

weight ×14

Hover or pick 0–5 to preview what each rating means.

Follows clear rules, not judgment

weight ×8

Hover or pick 0–5 to preview what each rating means.

Saturation Readiness0 / 100

0/12 rated

S 0P 0D 0Readout ↓