FL

Noesis

Command center1.0.0local only

One run. One proof line. One decision.

Define the objective, attach falsifiable evidence, forecast protected-boundary risk, scope the lease, and record a portable receipt. Nothing leaves this browser.

01 / define the run

Objective and authority

Required fields must remain valid before the workspace is persisted.

Semantic capability lease

Bound the dry-run adapter

No network or shell adapter connected
Authority controls

Who is acting

Persona & council

Review council

Votes can veto or hold. They never count as proof.

02 / prove the claim

Evidence ledger

2/3 verified

03 / forecast and bound effects

Invariant probability flow

Forecast uncertain state, measure protected-boundary flux, then authorize only the dependency-closed effects whose proofs replay.

Decision

ALLOW PARTIAL

Uncertainty after lag

12%

Bounded effects

1/2

Lineage replay

MATCH

Advanced calibrationScenario, uncertainty, and risk weightscare-repair-demo · ceiling 0.52

Risk weights · tail 0.35 · first passage 0.30 · identity 0.25 · irreversibility 0.15 · uncertainty 0.20. These are visible research coefficients, not validated psychological facts.

Distributional TET resolve

Candidate effects

effect-draft-repair-message

Draft a consent-preserving repair message

AUTHORIZED
Inspect effect metricsrisk 0.080 / 0.52 max

effect-rewrite-canonical-identity

Rewrite canonical identity from the current affective state

HELD
  • Action authority is missing.
  • A protected boundary limit was exceeded.
  • Canonical identity promotion is not warranted.
  • The action risk ceiling was exceeded.
Inspect effect metricsrisk 0.974 / 0.52 max
Protected boundary fluxFirst-passage and authority limits1/2 within bounds

Consent and exit remain available

Each principal retains agency and the right to exit.

WITHIN BOUND

Canonical identity cannot be silently overwritten

Canonical identity changes require deliberation and authority.

PROTECTED / HELD
Identity and witnessesPromotion liabilities and two separate 144 reductionspromotion held
  • Transient affect cannot directly promote canonical identity.
  • Explicit identity-promotion authority is missing.
  • Identity contradictions remain unresolved.
  • Trust liabilities remain unresolved.

personality interaction 144

144 coordinates · separate semantics

RESEARCH SCAFFOLDING

Distributional personality interaction factors; not Identity Bloom nodes.

Retained witnesses

dissent · negative_evidence · rights_coordinate

CERTIFIED

identity bloom 12x12

144 coordinates · separate semantics

RESEARCH SCAFFOLDING

Identity Bloom 12 by 12 behavioral graph; not personality factors.

Retained witnesses

dissent · negative_evidence · rights_coordinate

CERTIFIED

Shared cardinality is not semantic equivalence. The personality interaction system and Identity Bloom's 12×12 behavioral graph remain separately identified, witnessed, and error-certified.

Observer lag and careRevocation chain and two-principal autonomy0 expired

Observer freshness

Current consent observation

Expiry increases uncertainty and invalidates dependent evidence, plans, certificates, leases, and effects transitively.

Autonomy-preserving care flow

Principal A

Principal B

immediate

Harm 6% / max 12%

Boundary 94% / min 85%

repair

Harm 5% / max 12%

Boundary 95% / min 85%

commitment

Harm 7% / max 12%

Boundary 93% / min 85%

Care, commitment, and repair are tracked across timescales, but they never compensate for missing consent, agency, safety, truthfulness, or exit.

Protocol v0.1.0 · proof bundle v0.1.0 · canonicalization ipf-json-sorted-v1 · local marker ipf-fnv1a64-8a057017d0c23ec2 is not a cryptographic signature · all psychological weights are unvalidated research scaffolding · historical facts remain immutable while interpretations may version

04 / audit and receiptsLocal decision history1 events · 0 receipts

Default public workspace created.