Home
What needs your attention, what to do next, and how honest our record is so far.
Analyse a decision
One decision at a time. The workflow will help you sharpen it.
Needs your attention
Calibration
Active decisions
New decision
Two launch domains are live; opportunity/purchase ships after these meet their accuracy gates.
1 · Frame the decision
2 · Establish reality
Enter each number AND say what kind of information it is. The engine refuses to run on unknowns — it will not invent values.
3 · Reference class
4–5 · Model & simulate
Seeded Monte-Carlo: 5,000 monthly cash paths from your inputs. Same inputs → same result, always. Full methodology.
Scenarios
What moves this result (sensitivity)
Why this result & what could make it wrong
Challenge — the case against this conclusion
Evidence, assumptions & reference class
Show calculation (reproducibility)
Decision Library
Every saved decision, its frozen forecast, and its status.
Outcomes
Forecasts are only worth anything if you report what actually happened. Resolution criteria were frozen when each forecast was made.
Accuracy
Your calibration record, computed only from resolved outcomes. No resolved outcomes, no numbers — we don't invent accuracy.
Public methodology
How forecasts are produced, what the known weaknesses are, and how to reproduce any result: LEXUN Accuracy Centre.
Settings
Appearance
Your data
Everything lives in this browser. Export gives you the full auditable graph (decisions, frozen forecasts, outcomes, audit trail).
System — governor & platform intelligence
Real health checks and improvement tasks generated from this browser's own usage data. Nothing here is simulated; when there is no data, it says so.
Improvement tasks
Platform memory
Experimental (Labs)
The previous LEXUN app contains earlier experimental tools that have not passed this platform's release gates (documented methodology, reference classes, outcome tracking). They remain available, clearly separated: open Labs.