Defines the mission clearly
The decision is written down before analysis starts: who owns it, what counts as success, and by when. Vague questions produce confident answers to the wrong problem.
Cross-Industry Explainer
KRYOS V6
A Cross-Industry Governed Intelligence Framework
Structure complex decisions. Govern evidence. Adapt across domains. KRYOS V6 works through five things: a clearly defined mission, evidence assessed for quality, deliberate contradiction testing, scenario analysis with visible tradeoffs, and controlled release logic that decides what may actually be issued.
Adaptable is not the same as universal: it does not automate judgement and it does not produce certainty.
Built as an explainer. Avoids unsupported claims. Distinguishes documented capability from extrapolation.
The system
Five portable stages
Select a stage to read its definition. The same five stages carry between sectors; the domain content does not.
The decision being made, written down with its owner, horizon, and success condition.
Sources registered with provenance and assessed for fitness against this specific question.
A deliberate search for material that would defeat the working conclusion.
Assumption sets compared side by side so tradeoffs are visible rather than implied.
Proceed, qualify, constrain, escalate, or block — matched to the strength of support.
Capability
What KRYOS V6 actually does
Six functions, stated plainly. Each is a discipline applied to reasoning, not a feature that acts on your behalf.
Organises evidence and source quality
Each source is registered with provenance and assessed for fitness against this specific question. Credibility in general is not the same as fitness here.
Tests contradictions and competing explanations
The framework searches deliberately for material that would defeat the working conclusion, rather than waiting for it to appear.
Models scenarios and tradeoffs
Options are compared under stated assumption sets, so the tradeoffs are visible. A scenario describes a possibility, not a forecast.
Supports bounded recommendations
Output carries its assumptions, evidence basis, and named unresolved items. It advises; it does not instruct or execute.
Constrains, escalates, or blocks weak conclusions
Where support is insufficient the framework is designed to fail closed — narrowing, escalating, or withholding rather than softening the wording.
Boundaries
What it is not
Not this
- not a magic answer machine
- not autonomous authority
- not proof that a modelled scenario is true
- not universal coverage
- not a substitute for human accountability
More accurate framing
- a governed decision framework
- advisory and evidence-led
- scenario-explicit and assumption-aware
- bounded by scope and source quality
- designed to fail closed when support is weak
Portability
How the framework travels across sectors
The examples below are generic and realistic, not customer stories. What transfers between sectors is the reasoning discipline, not the domain content.
Common decision problems
- Supplier and third-party risk review
- Market entry or exit option comparison
- Build, buy, or partner assessments
- Post-incident commercial reviews
How KRYOS structures it
- Record who is authorised to accept residual risk
- Separate the analysis owner from the decision owner
- Log the evidence used at the moment of the decision
Outputs it might produce
- A scoped mission statement with named decision owners
- An evidence register ranked by source fitness
- Scenario comparisons with explicit assumptions
- A release class: proceed, qualify, constrain, escalate, or block
Must remain human-owned
Commercial commitment, risk acceptance, contractual obligations, and communication to boards or regulators.
A structured comparison is not a forecast. Scenario output describes what would follow if stated assumptions hold, not what will happen.
Common decision problems
- Supplier and third-party risk review
- Market entry or exit option comparison
- Build, buy, or partner assessments
- Post-incident commercial reviews
How KRYOS structures it
- Record who is authorised to accept residual risk
- Separate the analysis owner from the decision owner
- Log the evidence used at the moment of the decision
Outputs it might produce
- A scoped mission statement with named decision owners
- An evidence register ranked by source fitness
- Scenario comparisons with explicit assumptions
- A release class: proceed, qualify, constrain, escalate, or block
Must remain human-owned
Commercial commitment, risk acceptance, contractual obligations, and communication to boards or regulators.
Common decision problems
- Policy option comparison
- Programme continuation or closure review
- Cross-agency prioritisation of limited funds
- Consultation evidence synthesis
How KRYOS structures it
- Authority limits must be documented before analysis begins
- Publishable audit trail for scrutiny and review bodies
- Distinguish official statistics from modelled estimates
Outputs it might produce
- An auditable statement of scope and legal authority
- Contradiction notes where evidence sources disagree
- Option sets with distributional tradeoffs surfaced
- An explicit list of unresolved questions before decision
Must remain human-owned
Political judgement, statutory interpretation, public consultation duties, and final policy adoption.
Common decision problems
- Grant prioritisation across competing programmes
- Partner due diligence
- Programme redesign after evaluation findings
- Safeguarding and duty-of-care reviews
How KRYOS structures it
- Trustee-level sign-off recorded against the release class
- Beneficiary voice logged as a distinct evidence source
- Conflicts of interest declared inside the evidence register
Outputs it might produce
- A prioritisation rationale traceable to stated criteria
- Named evidence gaps in beneficiary data
- Scenario views of funding shortfall and surplus
Must remain human-owned
Mission interpretation, ethical judgement, beneficiary relationships, and trustee accountability.
Common decision problems
- Research synthesis across conflicting literature
- Structured expert elicitation
- Pre-publication challenge of a headline finding
How KRYOS structures it
- Funder influence declared in the evidence register
- Separate the synthesis role from the publication decision
Outputs it might produce
- A contradiction map of where sources disagree and why
- Claim classification by directly supported, inferred, or unresolved
- Scenario questions for future research
Must remain human-owned
Scholarly interpretation, methodological choice, and responsibility for published claims.
Common decision problems
- Service redesign option appraisal
- Procurement of clinical technology
- Incident and never-event review structuring
How KRYOS structures it
- Clinical accountability stays with named clinicians
- Patient data handling governed outside the framework
- Regulatory obligations override framework output
Outputs it might produce
- A scope statement bounded by clinical governance
- Contradiction checks between guidance and local data
- Escalation triggers where safety evidence is insufficient
Must remain human-owned
Clinical judgement, patient consent, regulatory compliance, and duty of candour.
Common decision problems
- Regulatory change impact assessment
- Compliance control gap review
- Escalation of contested findings
How KRYOS structures it
- Legal privilege boundaries defined before evidence intake
- No output treated as legal advice
Outputs it might produce
- A defensible record of what was considered and when
- Counterevidence search results
- Constrained recommendations with named conditions
Must remain human-owned
Legal opinion, privilege management, and regulatory representation.
Common decision problems
- Intervention selection and continuation
- Resource allocation across sites
- Evaluating vendor efficacy claims
How KRYOS structures it
- Safeguarding and learner data rules sit outside the framework
- Equity impacts recorded as an explicit tradeoff dimension
Outputs it might produce
- A record of which efficacy claims are supported and which are not
- Scenario views of rollout at different scales
Must remain human-owned
Pedagogical judgement, learner welfare, and institutional accountability.
Common decision problems
- Incident review and corrective action selection
- Maintenance prioritisation
- Change approval under uncertainty
How KRYOS structures it
- Operational authority levels defined per change class
- Time-boxed reviews recorded with the evidence available at the time
Outputs it might produce
- Ranked corrective options with residual risk stated
- Explicit block or escalate signals when support is weak
Must remain human-owned
Operational command, safety authority, and live intervention.
Common decision problems
- Incident attribution reasoning
- Control investment prioritisation
- Threat report credibility assessment
How KRYOS structures it
- Attribution statements require named authorisation
- Source reliability tracked separately from claim confidence
Outputs it might produce
- Competing-hypothesis comparison with confidence bounds
- Explicit statement of what the evidence cannot show
Must remain human-owned
Response authority, disclosure, law-enforcement liaison, and attribution statements.
Process
The KRYOS workflow
Define the mission
Write the decision down before analysing it: owner, horizon, success condition.
Learn more about define the missionLock scope and authority
Record what is in scope and who is permitted to accept the residual risk.
Learn more about lock scope and authorityGather and rank evidence
Register each source with provenance and assess its fitness for this question.
Learn more about gather and rank evidenceSurface contradictions
Search deliberately for evidence that would defeat the working conclusion.
Learn more about surface contradictionsCompare scenarios and options
Model assumption sets and make the tradeoffs between options explicit.
Learn more about compare scenarios and optionsRelease, constrain, escalate, or block
Match the strength of support to what may be issued, and fail closed when weak.
Learn more about release, constrain, escalate, or block
Application
Try your own use case
Answer five questions and we generate a starting framework map on the builder page. It runs entirely in your browser and produces a structure to work from, not a conclusion.
Reading claims
Evidence, limits, and unresolved questions
Evidence-backed core
Statements that follow directly from how the framework is defined — its stages, its release classes, and the separation between advice and authority.
Supported inference
Reasonable conclusions that require an interpretive step, such as the expectation that domain-neutral steps transfer between sectors.
Unresolved / requires verification
Claims this site does not support, including measurable outcome improvement and the availability of any specific component in a given deployment.
Go deeper
Explore the framework
What Is KRYOS V6
The canonical definition, why it exists, and what makes it portable.
How It Works
The seven-part operating model, from mission definition to retest logic.
Framework Components
Supporting layers, with maturity marked core, optional, derived, or proposed.
Glossary
Plain-language and technical definitions for every term used here.
Map KRYOS V6 to your context
Start with the builder to generate a structured starting map, or request a guided mapping session with a person.