Scenario studio
Compare repair, replacement, leasing and a service against the same need, with their different costs and assumptions.
ReFu · Vision
Explore what could happen.
See possible futures clearly enough to make a considered choice today.
Explore ReFu
Product design in development. What you can explore today.
Mission
Turn a clear question and permitted evidence into inspectable scenarios, assumptions and forecasts; preserve revisions and compare them with eventual outcomes.
One of Skyzai’s seven product strategies, designed to contribute to outcomes for Me (individuals), We (organisations) and They (machine intelligence).
Rosetta agent row · Methodology
What follows, and by which rule?
ReFu makes the route from evidence and assumptions to a possible outcome inspectable, so a forecast can be tested and revised.
Objectives
- Make assumptions and alternatives visible.
- Explain what could change an assessment.
- Compare forecasts with outcomes, including misses and unresolved cases.
Strategy
Compare alternatives over a common horizon, expose assumptions and revise evidence-led forecasts without erasing history. Define the baseline, eligible population and resolution first; return NoAssessment when the forecast basis is insufficient. YES/NO event markets, UP/DOWN trend cycles and multi-outcome winner markets remain separately qualified proposed offerings.
Forecast workspace
Keep evidence, horizons, uncertainty, abstentions and revisions with an assessment.
Resolution and market design
Declare what counts as an outcome before it happens. Proposed event, trend and multi-outcome markets need their own terms and qualification.
KPIs
Proposed measures of progress
Measure reader understanding and correction effort; evaluate forecasts against a declared baseline after outcomes resolve. Report misses, abstentions and disputed or unresolved cases alongside successes.
These measures describe how to evaluate this product’s contribution. Each needs a defined scope, observation period and evidence.
- Targets
- Not set. Thresholds remain to be agreed for the proposed measures.
- Performance observations
- No measured results are recorded here for these proposed measures.
Proposed evaluation contracts. Product observations: Not recorded. Targets: Not set. The fictional ledger below supplies no product measurements.Forecast error against a reference
Resolution and abstention coverage
Reader understanding
Correction reach and effort
Launch evidence
Dated measurements of product launches, live integrations and published customer outcomes have not been recorded here. Missing evidence is not a measured zero.
- Released services
- Not recorded
- Live product integrations
- Not recorded
- Published customer outcomes
- Not recorded
Historical declaration · . Owner direction, 18 October 2026. An owner-specified zero-release baseline was recorded for planning. It is retained as source history and is not a measured product result.
Agents
The Rosetta agent row connects this product’s field of inquiry to occupations, skills and responsibilities. The occupations below describe specialist work; different people, models and tools may carry these capabilities within an explicit assignment.
Rosetta agent row. Product L3 · Methodology
Responsibility. Evidence-qualified possibilities and forecasts
Reasoning approach. Deductive testing: make assumptions explicit, check what follows from them and test rival explanations. Keep the inference, uncertainty and eventual outcome distinct.
Scenario analyst
Skills and responsibilities. Develops alternatives, checks comparable scope and tests the assumptions that separate them.
Forecast researcher
Skills and responsibilities. Examines the basis for an assessment and returns NoAssessment when evidence cannot support it.
Outcome reviewer
Skills and responsibilities. Applies a fixed resolution rule to independent evidence; preserves misses, unresolved cases and disputes. Market operation needs separate qualification.
Required capabilities include forecast authorship, domain review, independent outcome resolution, evaluation and product engineering. Actual appointments and capacity remain separate.
Required roles, not staffed appointments or permission grants. The L3 Methodology correspondence describes a product’s inquiry; it is not a technical network layer.Scenario analyst · input and output
Forecast researcher · input and output
Independent outcome reviewer · input and output
These are roles and skills in development, with a defined context and an output to inspect. Their performance needs evidence and evaluation; a role description does not give an agent permission to act.
See it in a piece of work.
Local example · fictional workshop, costs, probabilities and outcome records. Nothing is sent or retained after leaving or reloading this page. No supplier, model, market or payment service is connected.
Repair, replace, lease or buy the service?
Nora’s workshop needs Workcell 17 contributing again. Before its separate actuator purchase, compare four ways to meet an internal inspection need. Restoring equipment does not qualify the workshop’s later customer service.
- Question and horizon
- Will at least 95 of 100 scheduled internal inspection hours be completed during the next 90 days? Day 0 is a separately agreed start; day 90 is the fixed review point.
- Common scope
- The same 100 scheduled hours, tooling, safety and reporting requirements. Costs are invented direct charges in USD over three months; tax, downtime, energy, financing and exit costs are unknown. These are not comparable full-cost quotes.
- Scenario comparator
- No intervention: the robot remains unavailable. The workshop would need another way to cover the inspection schedule. This comparator is not a measured base rate.
- Resolution rule, fixed before the outcome
- Count completed hours from an independent inspection log at day 90, using the schedule frozen at day 0. At least 95 hours means Yes; fewer means No. Missing or contested logs remain unresolved or disputed. Changing the schedule requires a new question. Equipment availability alone is insufficient.
Inspect the basis
Fictional source sheet v1 assumes a 12-day repair part lead time. Its correction, v2, changes that to 28 days. The numbers illustrate an assessment revision; no statistical model or historical fit produced them. Service capacity has no supporting source in either version.
Assessment 1: Based on fictional source v1 and its stated assumptions.
Repair
Replace the actuator in the existing robot.
- Illustrative direct cost
- USD 450 once
- Decisive assumption
- A qualified part fits, arrives in 12 days and passes installation and safety inspection.
70% conditional forecast
Conditional on those assumptions. The outcome remains unresolved; alternatives are not independent bets.
Replacement
Obtain and qualify another robot.
- Illustrative direct cost
- USD 1,600 once
- Decisive assumption
- Delivery in 10 days, compatible tooling, training and a completed safety inspection.
85% conditional forecast
Conditional on those assumptions. The outcome remains unresolved; alternatives are not independent bets.
Leasing
Use a leased robot for the 90-day comparison.
- Illustrative direct cost
- USD 95/month; USD 285 for three months
- Decisive assumption
- A three-month term is available, servicing is included and tooling is compatible. Return terms need review.
80% conditional forecast
Conditional on those assumptions. The outcome remains unresolved; alternatives are not independent bets.
Contracted service
A provider performs the internal inspections.
- Illustrative direct cost
- USD 300/month; USD 900 for three months
- Decisive assumption
- Provider capacity, access, reporting and acceptance terms are not supplied.
NoAssessment · insufficient basis
Withhold a likelihood and obtain the missing evidence before preferring this path.
No path is preferred. A preference grants no purchasing, leasing, trading or payment authority. Agentz can separately help prepare an unsigned proposal; the workshop must qualify terms, authority and acceptance. No earlier assessment in this page session. Returning to v1 creates another assessment; it does not erase a correction or make v1 current evidence. This separate teaching cohort contains five distinct workshop assignments, each with its own realised path. All used the question and resolution rule above, with 90-day horizons ending in September 2026. The five records, issue-time probabilities and reference predictor were frozen before their outcomes. They are not the four unchosen futures of Workcell 17. Issued Yes probability: 80%. Outcome: Yes. Fictional independent reviewer: 97 completed hours; signed-off log supplied at day 90. Issued Yes probability: 70%. Outcome: No. Fictional independent reviewer: 82 completed hours; the forecast missed. Issued Yes probability: 60%. Outcome not assigned. Day 90 reached; the independent log is missing. No outcome assigned. Issued Yes probability: 75%. Outcome not assigned. Two inspection logs disagree. Independent resolution remains open. Abstained at issue. Outcome not assigned. NoAssessment at issue: the available basis did not support a probability. Registered: 5. Scored: 2. Due unresolved: 1. Disputed: 1. Abstained: 1. Resolution coverage: 2 of 4 issued forecasts; abstention: 1 of all 5 registered assignments. Reference: a fixed 50% Yes predictor for this teaching cohort, not a historical base rate. On the identical two qualified resolved forecasts, mean squared probability error (Brier score) is 0.265 versus 0.250 for the reference; lower is better. This example underperforms the reference. The miss remains included. Unresolved, disputed and abstained records stay unscored and visible; the small selected resolved subset proves no calibration or product performance.Assessment history (0 earlier records)
How forecasts meet outcomes
Inspect the separate actuator proposal · Continue with an unsigned Agentz proposal · Review the source accounts in Circle
A separate analysis-commissioning example
Illustrative case · Help our team choose and commission an analysis within our budget.
Consider more than one way forward.
With the question and its sources in view, the team explores alternatives before choosing an approach.
- Quicker analysis
- May help the immediate decision; depends on access to the necessary material.
- Independent review
- May offer a different perspective; time, cost and confidentiality need comparison.
- What could change
- New evidence, a changed deadline or a different budget could change the assessment.
A fictional example of the intended experience. No request is sent, personal memory stored or money moved.
Carry the question forward with AgentzOne job, with clear boundaries.
- Starts with
- A clear question, permitted evidence, a time horizon and explicit assumptions.
- Produces
- Inspectable scenarios, forecasts, revisions and an explicit insufficient-basis result when an assessment cannot be supported.
A forecast is not a promise, a recommendation or a transaction. Missing evidence remains missing; it does not become an invented probability.
Capabilities and current proof.
As reviewed on , these are selected capabilities traced to their product owner. The proof level describes a plan, source observation or bounded test; it does not establish a released service.
Attributed forecast/market source · Bounded local proof
A local browser read a real public ReFu market response and retained its status and attribution.
Current limit: A price/status observation is not a verified forecast, tradable market, current outcome or service acceptance.
Forecast horizon and resolution · Proposed
The owner assigns explicit horizons and resolution rules to Reality Futures.
Current limit: No accepted question-to-outcome lifecycle is evidenced by this trace.
Optional market formats · Proposed
YES/NO, UP/DOWN and multi-outcome formats are noted as future qualification candidates.
Current limit: A market format is not an operating market or permission to trade.
Product plan, responsibilities and sources
Direction and delivery
Skyzai’s vision is an emergent Marketsphere. Its mission is to develop SPECTRE as an open operating system for it. Me, We and They describe the participant objectives, and each of the seven products is a strategy with its own KPIs and Agents. This nested product plan explains its direction, objectives, delivery work, measures and responsibilities; customer outcomes remain to be demonstrated.
- Vision
- People can examine possible futures, their reasoning and uncertainty before making consequential choices.
- Mission
- Turn a clear question and permitted evidence into inspectable scenarios, assumptions and forecasts; preserve revisions and compare them with eventual outcomes.
- Objectives
- Help people understand alternatives, identify decisive assumptions, recognize missing evidence and see what would change an assessment. Make forecasts independently testable.
- Strategies
- Compare alternatives over a common horizon, expose assumptions and revise evidence-led forecasts without erasing history. Define the baseline, eligible population and resolution first; return NoAssessment when the forecast basis is insufficient. YES/NO event markets, UP/DOWN trend cycles and multi-outcome winner markets remain separately qualified proposed offerings.
- Measures & triggers
- Measure reader understanding and correction effort; evaluate forecasts against a declared baseline after outcomes resolve. Report misses, abstentions and disputed or unresolved cases alongside successes.
- Agents & skills
- Required capabilities include forecast authorship, domain review, independent outcome resolution, evaluation and product engineering. Actual appointments and capacity remain separate.
Plan basis and open decisions
Current owner direction defines product responsibilities. Retained proposals inform delivery plans and engineering work. Signed charters, current operating mandates, staffing and measured baselines retain their separate standing.
- The customer purpose is a scenario-and-forecast workspace. Its packet contract is supporting infrastructure, and reader understanding needs its own proof.
- Prediction markets have separate economic, custody and outcome-resolution requirements. A market price remains distinguishable from a forecast; newer charter wording remains proposed.
These working documents require access to the corresponding private GitHub repositories.
- Adopted owner product direction · 2026-10-13
- Reviewed product proposal and retained delivery plan · 2026-10-09
- Corresponding Rosetta L3 row · 2026-10-09
Rosetta agent row.
Selected product correspondence. ReFu occupies L3: what follows, and by which rule? Its forecasts preserve reasons, rival explanations, horizons and uncertainty.
- Product L3 · Methodology
- Evidence-qualified possibilities and forecasts
What follows, and by which rule? ReFu makes the route from evidence and assumptions to a possible outcome inspectable, so a forecast can be tested and revised.
How the Rosetta agent row is used
Skyzai presents seven products: APU, Circle, ReFu, Agentz, Menexus, LIQUID and MurMur. Agentz is the continuing centre, with Cortex memory and Nexus connections. APU is the interface. Menexus means ensurance, insurance and assurance. Product roles preserve the underlying Rosetta coordinates. Emergentism is the framework across the whole. The Spectre network is a component within the wider Marketsphere programme; Permanent Composability supports independent reuse across the products.
The Rosetta agent row relates this product’s inquiry, reasoning and specialist skills to Emergentism. Several models, people or tools may contribute to the work. The philosophical correspondence, product responsibility and agent dispatch contract remain distinct; this page neither establishes deployed performance nor grants a model permission to act. This product direction does not migrate runtime contracts or amend signed source history.
Explore the design.
These pages describe the intended product and its delivery work. They do not open a signed-in app or establish operating availability.