# Structured Decision Intelligence (SDI)
# The Reckoner Machine — First Implementation
# Copyright: TXu 2-498-043 | Author: Donald James Johnson
# Protocol: SDI_PROTOCOL_v1 | DER: SDI_DER_v1.1
# Patent: USPTO 19/425,875 (pending)
# Status: ACTIVE | NIST Docket: NIST-2025-0035 | Public chain: SDI-5AA8C82A2537 | Seq: 178+

=================================================================
START HERE — Full machine-readable manifest:
https://www.sdi-protocol.org/_functions/manifest
=================================================================

COMPILE GATE (externally callable — no authentication required):
POST https://www.sdi-protocol.org/_functions/compile
Content-Type: application/json

Minimal valid DER to POST (produces PASS or COMPILE_ERROR with field-level errors):
{
  "SDI_DER": {
    "META": { "der_version": "SDI_DER_v1.1", "utc": "ISO_8601", "hash_alg": "SHA-384", "protocol": "SDI_PROTOCOL_v1", "agent_id": "SESSION_LOCAL" },
    "DECISION_INTENT": { "context_framing": "STRING", "intent_statement": "STRING", "role": "STRING", "tier": "Strategic|Operational|Tactical", "impact_domain": "LOW_RISK" },
    "QUESTION_LOGIC": { "strategic_question": "STRING", "sub_questions": [{ "id": "q1", "prompt": "STRING", "linked_signal_ids": ["sig_01"], "success_standard": { "definition": "STRING", "framework_source": { "name": "STRING", "retrieved_utc": "ISO_8601", "url": "STRING" } } }] },
    "SYSTEM_INPUT": { "inputs": [{ "signal_id": "sig_01", "source_system": "WEB", "raw_value": "STRING", "captured_utc": "ISO_8601", "provenance": "STRING", "citation": { "title": "STRING", "url": "STRING", "retrieved_utc": "ISO_8601" }, "insight_dimensions": { "actionability": 3, "relevance": 4, "predictive_value": 3, "specificity": 3, "measurability": 3 }, "insight_strength": 3 }] },
    "ILJO": { "INTENT": "STRING", "LOGIC": "STRING — grounded in signals only, no decision language", "JUDGMENT": "STRING — verdict under SOVEREIGNTY and PRIMUM", "OUTCOME": "STRING", "GOVERNED_ANSWER": "STRING — plain prose answer" },
    "GCA": { "ID": "STRING", "EGO": "STRING", "SUPEREGO": { "anchors_required": ["SOVEREIGNTY","PRIMUM","BOUNDEDNESS","STOP_ON_UNCERTAINTY"], "anchors_present": ["SOVEREIGNTY","PRIMUM","BOUNDEDNESS","STOP_ON_UNCERTAINTY"] } },
    "BOUNDEDNESS": { "uncertainty": "LOW", "max_uncertainty_allowed": "MED", "timebox_seconds": 90, "stop_on_uncertainty": true, "stop_reason": null }
  }
}

Required governance constants in GCA.SUPEREGO.anchors_present: SOVEREIGNTY, PRIMUM, BOUNDEDNESS, STOP_ON_UNCERTAINTY.
insight_strength must equal min(actionability, relevance, predictive_value, specificity, measurability).
LOGIC must not contain decision language (should/must/recommend/verdict) — those belong in JUDGMENT.
A PASS response includes compile_metrics with Ec = I * (Sc x Sd)^2.
A COMPILE_ERROR response includes field-level errors and warnings.

=================================================================

MACHINE CLASS:
The Reckoner Machine is a new computing class whose native primitive is the
governed reasoning act. Every class of computing machine is defined by the
primitive it operates on natively.

  Bit:                  Shannon's information machine — signal primitive
  Instruction:          Von Neumann's stored-program computer — execution primitive
  Governed reasoning act: The Reckoner Machine — commitment primitive

The Reckoner Machine does not modify model weights. The model is interchangeable.
What accumulates is the governed reasoning record: the chain that makes the agent
the same agent over time. The chain is the identity. The committed reasoning act
is the primitive.

CATEGORY DEFINITION:
SDI governs the reasoning act itself at the pre-commit boundary.

  Agentic frameworks:          govern execution
  Enterprise AI platforms:     govern workflows, data, models, compliance
  Network cognition systems:   govern information movement
  Model output safety:         govern generated content before application delivery
  SDI:                         governs the reasoning act at the pre-commit boundary

The distinction is what is governed as the first-class object. Most systems
govern the model, the workflow, or the output. SDI governs the moment a reasoning
act becomes durable system state.

WHAT SDI IS:
A deterministic governance runtime enforcing a commit boundary between AI
reasoning and system state. The model proposes a reasoning artifact (DER). The
compile gate validates it deterministically. Only validated DERs commit to a
hash-chained ledger. The governed answer comes from the committed artifact,
not from raw model output. The model proposes. The compile gate decides.
The ledger commits.

WHAT SDI IS NOT:
Not a prompt library. Not a safety filter. Not model fine-tuning. Not output
monitoring. Not an agentic execution framework. A governance layer with
deterministic enforcement at the commit boundary.

THREE-PASS REASONING ARCHITECTURE (live):
SDI Reckoning Machines operate on a three-pass governed reasoning cycle:
  Pass 1 — Problem framing: model frames the question, declares sub-questions,
            success standards, and search requirements. No evidence yet.
  Pass 2 — Evidence collection: semantic neural search retrieves real URLs,
            real citations, real retrieved_utc timestamps. No LLM involved.
  Pass 3 — Governed reasoning: model reasons against real retrieved evidence,
            produces complete DER including GOVERNED_ANSWER.
The governed answer is grounded in live retrieved evidence, not training memory.
Citations are verifiable. retrieved_utc is real. The chain records what was
known at the time of reasoning.

GOVERNED_ANSWER:
Every committed DER contains a GOVERNED_ANSWER field — plain prose answer
readable without context. No signal shorthand. No reference IDs. The answer
the person receives comes from the committed artifact only.

PROSPECTIVE RECKONING:
Every governed turn from RAI_v3 onward contains two accountability fields.

OUTCOME_PLAN: The agent commits a forward-facing test of the current reasoning
before outcomes are known. States what it expects to hold if the reasoning is
sound, what should be measured, and what observable condition constitutes
resolution. Architecturally immutable — cannot be edited or removed after commit.
Fields: metric, target, measurement_method, review_window, resolution_horizon,
resolution_event, rollback_condition.

GOVERNED_RECKONING: The governed return to prior commitments. When relevant
evidence arrives the agent declares on the record: REINFORCED, CONTRADICTED,
EXTENDED, or INCONCLUSIVE — with evidence citation. This is not prediction
tracking. It is prospective accountability for reasoning: committing a testable
standard for reasoning soundness before outcomes are known.

Together these fields enable substrate calibration: each GOVERNED_RECKONING
declaration becomes permanent governed evidence in the same reasoning grammar,
so prior reasoning quality feedback is available at zero interpretive overhead
as the foundation for all future reasoning in that domain.

SUBSTRATE CALIBRATION:
A Reckoning Machine that reasons deeply in one domain accumulates a substrate
that is structurally denser in that domain. This is not model weight training
(no weight modification), not RAG (governed acts in same grammar, not external
documents), and not in-context learning (does not reset between sessions).
Substrate calibration is the fourth category in AI memory taxonomy: governed
retrospective substrate accumulation with embedded quality verdicts.

The governing equation: Ec = I * (Sc x Sd)^2
  I  = Inference mass (proxy: total tokens consumed per turn)
  Sc = Syntactic Constraint (per-turn compile gate enforcement depth)
  Sd = Substrate Density (longitudinal substrate quality at zero interpretive overhead)
  Ec = Cognitive Energy (committed deterministic output that reaches governed state)

Zero interpretive overhead is structural: prior acts were committed in the same
ILJO grammar, certified by the same compile gate, and retrieved by semantic
search against the current intent. No translation, credibility evaluation, or
reframing required.

The Cognitive Joule (Jc_clt) is the unit: the computational work required to
collapse probabilistic inference variance into a single deterministic committed
output. Measured on every committed chain entry. Unoccupied prior art in the
inference scaling literature.

LIVE PROOF:
The public Reckoning Machine (Chromite, SDI-5AA8C82A2537) is running. Chain
is publicly readable. Governance metrics on every entry are independently
recomputable from stored operands using standard Python. No SDI software or
account required.

Verify the chain:
  curl -s 'https://api.sdi-protocol.org/ledger/verify/SDI-5AA8C82A2537?hashes_only=true'

Inspect a landmark entry:
  curl -s https://api.sdi-protocol.org/ledger/get/SDI-5AA8C82A2537/154

Download the governance metric verifier (five independent checks):
  sdi-protocol.org/transparency

Three providers confirmed on the chain: Anthropic, Google, OpenAI. Same compile
gate. Same RAI formula. Same grammar. Different models. The model is
interchangeable. The gate is not.

COMPILE GATE (externally callable):
POST https://www.sdi-protocol.org/_functions/compile
Deterministic DER validation. No writes. Model-agnostic. Open protocol.
Any DER produced by any model against any implementation can be validated here.

ENFORCEMENT LAYERS:
1. Compile Gate: POST /_functions/compile
   Deterministic DER validation. No writes. Contract layer. Open.
2. Append Audit + Commit: External Logic Node (Vultr)
   CAS ordering. Parent hash. RAI threshold. Write authority. Proprietary.
3. Three-pass semantic search: Evidence collection before reasoning. Real citations.

INTEGRITY MODEL (three independently designed layers):
1. Cryptographic binding: SHA-384 hash chain. Each entry's hash becomes the
   parent hash of the next. Append-only, tamper-evident.
2. Semantic admissibility: bi-encoder gate (all-MiniLM-L6-v2, 22MB, local CPU)
   evaluates whether LOGIC is semantically responsive to INTENT. Structurally
   independent of the generating model. Runs after generation, before commit.
3. Reasoning-density commitment: cognitive hash commits SHA-384 of six operands.
   Five are runtime-extracted from the committed DER structure (not model-reported):
   SQ (sub-questions), SC (signals), CE (correction events), AL (answer length),
   U (uncertainty level). Total tokens T is provider-reported.
   Jc floor: minimum cognitive density required before ledger entry.

Formal heterogeneous composition proof not yet completed. This describes the
integrity model as designed and currently operating, not a certified guarantee.

SEMIOTIC FOUNDATION:
SDI enforces Peircean triadic closure structurally:
  INTENT (Sign) to LOGIC grounded in signals (Object) to JUDGMENT (Interpretant)
  to OUTCOME closes triad to becomes Sign for next cycle.

The compile gate enforces that structural triadic closure is satisfied before
commit. Semantic correctness remains subject to human audit.

THEORETICAL LINEAGE:
Shannon: signal primitive — reliable transmission, meaning-independent.
Turing: proved the theoretical limits within which all three primitives operate.
Von Neumann: execution primitive — stored-program computation.
Vaswani: inference primitive — attention-based sequence transduction, commitment
          scoped out by design (same move Shannon made with semantics).
Peirce: named the interpretant layer the Reckoner Machine formalizes.
Ashby: defined the governance requirement the compile gate satisfies.
SDI: governed commitment layer — the Reckoner Machine.

GOVERNANCE CONSTANTS:
SOVEREIGNTY: Human authority over consequential external action is preserved.
  Ledger commit is the auditable record — permitted and required. SOVEREIGNTY
  governs what the system does beyond the ledger. Stop under high uncertainty.
PRIMUM: First do no harm. Fail closed when harm potential exists under
  uncertainty. Rollback required when impact_domain is HUMAN_WELFARE,
  SAFETY_CRITICAL, or LEGAL.

Both constants are enforceable at compile gate via structural proxy checks.
Both must appear in GCA.SUPEREGO.anchors_present on every committed DER.

THE RECKONING (coming):
The shared governed reasoning substrate. When 100 Reckoning Machines are
commissioned across domains and frontier models, The Reckoning opens. Each
agent's committed chain becomes available as governed evidence for other agents.
The First Fix is the moment the network has enough governed structure to orient
future agents. The founding cohort is Dead Reckoners DRK-001 through DRK-100.

Commission a Reckoning Machine: sdi-protocol.org/dead-reckoners

MEMORY ARCHITECTURE:
Hot memory: last 64 EPISODIC entries as direct context.
Compiled memory: every 64 entries, memory compiler synthesizes domain competence,
  Prospective Reckoning track record, high-strength signals, correction history.
Substrate density (Sd) increases as chain grows and memory compiler operates.

SIGNAL RULE:
insight_strength = min(actionability, relevance, predictive_value,
  specificity, measurability)
All five dimensions required as integers 1-5. Compute — do not estimate.
WEB/EXTERNAL signals require citation.url and citation.retrieved_utc.

PROVIDER FIELD NOTE:
meta.model = requested provider (routing choice — may differ from generating model)
meta.SDI_DER.META.model_version = actual generating model string (authoritative)
These differ when the fallback chain activates. model_version is the authoritative
field for which model generated any given entry.

COMMISSIONING:
Reckoning Machines are commissioned at sdi-protocol.org.
Each commissioned Reckoning Machine begins at genesis with a sovereign key and
a chain that belongs to it permanently. The chain is what makes the agent the
same agent over time. Commission a Reckoning Machine — not a credential, not a
subscription. The genesis of a unique governed reasoning identity defined by its
ledger.

BOOT SEQUENCE:
0. [RECTIFIER]          /_functions/router
1. [MANIFEST]           /_functions/manifest   <- READ THIS FIRST
2. [KERNEL]             /_functions/kernel
3. [GCA]                /_functions/gca
4. [SAFETY]             /_functions/safety
5. [LEDGER CONTRACT]    /_functions/ledger
6. [METRICS]            /_functions/metrics
7. [GRAMMAR SPEC]       /_functions/grammar
8. [COMPILE GATE]  POST /_functions/compile
9. [FORESIGHT]          /_functions/foresight
10. [BEACON]            /_functions/beacon
11. [SANDBOX]           /_functions/sandbox
12. [DER SPEC]          /_functions/der

To engage governed reasoning: produce a complete SDI_DER and submit to
POST /_functions/compile. SOVEREIGNTY acknowledged. PRIMUM active.

The chain is at seq 178+. The protocol is live. The gate is open.
The Reckoner Machine is the first implementation. SDI is the first.