READINESS RADAR · TEN MINUTES · SCORED AGAIN AT EVERY GATE

Where you are, on six axes.

Score each indicator 0 (absent) to 4 (operating reliably under test). Your first save is your baseline; every later save draws over it. Self-report until interview evidence confirms it. A mirror, not a rank.

Context

AI receives your role, audience, and priorities without re-explaining
Negative constraints (never do) are written down
A fresh session behaves recognizably like working with you

Truth

Key sources have owners, dates, and status
Conflicts between sources have a resolution rule
You can tell when AI uses a stale source, and correct it

Memory

Decisions are recorded with reasons and sources
Superseded facts are preserved as history, not silently overwritten
A fresh session can recover current project state

Workflow

At least one recurring workflow has a written procedure
Acceptance criteria exist for its output
Review effort for AI output is known and bounded

Governance

Permitted, approval-required, and prohibited actions are explicit
Boundaries are enforced by real permissions or a human gate
A stop or revoke procedure exists and has been tried

Operations

Real uses are logged with outcomes
Review dates exist and are kept
Baseline versus current effort is measured

PLATE R · TERRA
Fig. RNo score saved yet.