Go / No-Go Gate
Experimental
Readiness Check
Check one story, epic, or business ask against five readiness gates.
Your assessment will appear here.
The standard
How the scoring works
Every item is checked against the same five gates. Only information stated in the item receives credit. Clear writing can still leave important decisions unanswered.
Scores
| Score | Label | Meaning |
|---|---|---|
| 0 | Missing | The item does not address this gate. |
| 1 | Needs clarification | The gate is addressed, but the team would still need to guess or ask. |
| 2 | Clear | The description provides usable information for this gate. |
Verdict
| Scores | Verdict | Why |
|---|---|---|
| Any gate is 0 | Not ready | At least one gate is missing. |
| No 0s, at least one 1 | Needs work | All gates are addressed, but some still need clarification. |
| Every gate is 2 | Ready | All five gates meet the clarity standard. |
The verdict is calculated from the five scores by the fixed rule above, not chosen by the AI. There is no total out of ten: a strong gate cannot make up for a missing one. A strong write-up cannot talk its way into a pass.
The tool checks the description, not the world. It does not verify claims, judge whether the idea is valuable, or confirm that the work is feasible, approved, or safe to release. Confirm the findings with your team.