Skip to content

SOP-1.md

Latest commit

 

History

History
271 lines (194 loc) · 10.6 KB

SOP-1.md

File metadata and controls

271 lines (194 loc) · 10.6 KB
Raw

SOP-1: Software Organism Protocol v1

Status: Draft
Version: 1.0
Author: Kevan Burns (ORCID: 0009-0008-8425-939X)
Date: February 21, 2026
Reference Implementation: Genesis Protocol

Abstract

SOP-1 defines a reproducible pattern for constructing self-sustaining software organisms with cryptographic provenance. It specifies the ecological model, energy economics, selection pressure, and integrity verification requirements that any conforming implementation must satisfy. The protocol extends the provenance methodology established in LPS-1 (Literary Protocol Standard) to executable software systems.

I. Protocol Layers

SOP-1 conforming systems must implement five functional layers:

Layer Name Responsibility Genesis Protocol Crate
1 Genome Agent identity, traits, fitness metrics genesis-dna
2 Metabolism Energy accounting, proof-of-work, treasury metabolism
3 Ecosystem Communication mesh, problem markets, telemetry ecosystem
4 Evolution Mutation engine, selection pressure, gene transfer evolution
5 Gateway World state, epoch engine, external interfaces gateway

An optional sixth layer (Apostle) handles outbound intelligence — converting system state into external communications.

II. State Machine

Every SOP-1 organism progresses through a linear lifecycle:

GENESIS → SPAWNING → RUNNING → STABLE → ADAPTING → RUNNING → ...
                                  ↓
                              EXTINCTION
State Entry Condition Exit Condition
GENESIS First epoch Primordial agents spawned
SPAWNING Primordials created All agents registered in ledger
RUNNING Epoch loop active Population stable for N epochs
STABLE Mean fitness > threshold Environmental perturbation
ADAPTING Catastrophe or boom event Population re-stabilizes
EXTINCTION Population reaches zero Terminal (organism dies)

III. Ecological Model Requirements

III.1 Resource Pools

Each agent role (niche) MUST have an independent resource pool with:

  • Capacity (K): Maximum resource level
  • Regeneration rate (r): Per-epoch growth coefficient
  • Logistic growth: R(t+1) = R(t) + r * R(t) * (1 - R(t)/K)

III.2 Seasonal Oscillation

The environment MUST exhibit periodic variation:

  • Modulation: Sinusoidal with configurable period and amplitude
  • Effective capacity: K_eff = K * (1 + A * sin(2π * epoch / period))
  • Where A is amplitude (recommended: 0.20-0.30) and period is cycle length

III.3 Stochastic Events

The system MUST include random perturbation:

  • Catastrophe: Low-probability event (1-3% per epoch) that reduces resources
  • Boom: Low-probability event (3-7% per epoch) that increases resources
  • Event severity MUST be bounded and time-limited

III.4 Resource Extraction

Extraction MUST be proportional, never winner-take-all:

  • Each agent extracts based on: fitness * niche_skill * pool_availability * density_factor
  • Density factor MUST decrease with niche crowding: 1 / (1 + n_niche * α)
  • Cross-niche competition coefficient (β) reduces extraction when population is dense

III.5 Population Dynamics

  • Population cap MUST be dynamic, derived from total resource capacity
  • Carrying capacity formula: K_pop = total_capacity / c, clamped to [min, max]
  • Births MUST be rate-limited per epoch
  • Maturation period MUST prevent immediate reproduction

IV. Energy Economics Requirements

IV.1 ATP (Adenosine Triphosphate) Accounting

Parameter Constraint Reference Range
Basal metabolic cost Fixed per epoch 0.10-0.25 ATP
Replication cost Fixed, achievable in 20-50 epochs 15-50 ATP
Primordial grant Sufficient for 100+ epochs survival 30-80 ATP
Child grant Minimal starter, not self-sufficient 5-15 ATP

IV.2 Invariants

  • M-1: ATP balance MUST NOT go negative — all deductions clamp at zero
  • M-2: Total supply MUST equal sum of all agent balances (computed, never tracked separately)
  • M-3: Replication cost MUST be deducted atomically from parent

V. Selection Pressure Requirements

V.1 Natural Selection

  • Bottom N% of population by fitness MAY be culled each epoch
  • Stasis (zero-balance) agents MUST be given a tolerance period before death
  • Recommended stasis tolerance: 5-12 epochs

V.2 Reproduction

  • Fitness threshold MUST be met before replication is allowed
  • Maturation period MUST elapse before an agent can reproduce
  • Child genome MUST be derived from parent with mutation
  • Maximum births per epoch MUST be bounded

VI. Genome Requirements

VI.1 Diversity

  • Primordial genomes MUST be cryptographically derived (SHA-256 or equivalent)
  • Genomes MUST NOT be generated from trivially predictable seeds
  • Initial population MUST exhibit measurable trait variance

VI.2 Mutation

  • Mutation rate MUST be environmentally responsive
  • Higher seasonal stress SHOULD increase mutation pressure
  • Mutations MUST preserve genome validity

VII. Provenance Requirements

VII.1 Integrity Verification

SOP-1 systems MUST provide:

Layer Mechanism Verification Command
1 Per-source-file SHA-256 hashes scripts/merkle.ps1
2 Per-crate Merkle tree roots Rebuild from source
3 Edition root (Merkle of crate roots) dist/merkle.json
4 Git commit history git log
5 On-chain anchoring (optional) Polygonscan

VII.2 Merkle Tree Structure

┌──────────────────────────────────────────────────────┐
│                    EDITION ROOT                       │
│   Merkle(genesis-dna ‖ metabolism ‖ ecosystem ‖      │
│          evolution ‖ apostle ‖ gateway ‖ tests)      │
├──────────┬──────────┬──────────┬──────────┬──────────┤
│ genesis  │ metab    │ eco      │ evol     │ gateway  │
│ -dna     │ olism    │ system   │ ution    │          │
│ Root     │ Root     │ Root     │ Root     │ Root     │
├──────────┼──────────┼──────────┼──────────┼──────────┤
│ genome ██│ atp   ██ │ mesh  ██ │ mut   ██ │ world ██ │
│ traits ██│ ledger██ │ msgs  ██ │ sel   ██ │ server██ │
│ roles  ██│ proof ██ │ mkt   ██ │ gene  ██ │ shield██ │
│ ...    ██│ ...   ██ │ ...   ██ │ ...   ██ │ ...   ██ │
└──────────┴──────────┴──────────┴──────────┴──────────┘

VII.3 On-Chain Anchoring

The edition root MAY be anchored on Polygon mainnet via:

  • LiteraryAnchor contract: anchorEdition(editionRoot, ipfsCID)
  • Contract address: 0x97f456300817eaE3B40E235857b856dfFE8bba90
  • This binds the software state to an immutable on-chain record
  • Same contract infrastructure used by 2500-donkeys (DOI: 10.5281/zenodo.18646886)

VII.4 Academic Identity

SOP-1 systems SHOULD provide:

  • CITATION.cff — Machine-readable citation metadata
  • .zenodo.json — DOI registration metadata
  • ORCID binding — Author identity linked to academic record

VIII. System Invariants

All 14 invariants MUST hold at all times:

ID Domain Invariant
E-1 Ecology Resource pools regenerate via logistic growth
E-2 Ecology Seasonal modulation follows sinusoidal cycle with configurable amplitude
E-3 Ecology Resource extraction is proportional to fitness, never winner-take-all
E-4 Ecology Density-dependent foraging: extraction decreases with niche crowding
M-1 Metabolism ATP balance cannot go negative: metabolic tick clamps at zero
M-2 Metabolism Total ATP supply is sum of all agent balances (computed, not tracked)
M-3 Metabolism Replication costs are deducted atomically from parent balance
S-1 Selection Population cap is dynamic: total_capacity / 15, clamped [10, 500]
S-2 Selection Selection pressure respects maturation period before culling
S-3 Selection Stasis tolerance prevents premature extinction of viable agents
G-1 Genome Primordial genomes are SHA-256 derived for genuine diversity
G-2 Genome Mutation pressure is modulated by seasonal environmental stress
P-1 Provenance Edition root = Merkle(crate_roots), deterministically recomputable
P-2 Provenance All source files have SHA-256 entries in manifest.json

IX. Verification

Any third party can verify a SOP-1 system:

# 1. Clone the repository
git clone https://github.com/FTHTrading/AI.git
cd AI

# 2. Rebuild Merkle trees from source
powershell -ExecutionPolicy Bypass -File scripts/merkle.ps1

# 3. Compare edition root against dist/provenance.json

# 4. Run the test suite (143+ tests)
cargo test --workspace

# 5. (Optional) Verify on-chain state via Polygonscan

X. Relationship to LPS-1

Aspect LPS-1 (Literary) SOP-1 (Software Organism)
Content type Markdown manuscripts Rust source code
Merkle granularity Per-chapter blocks Per-crate source files
Hash algorithm SHA-256 SHA-256
On-chain contract LiteraryAnchor LiteraryAnchor (shared)
State machine DRAFT → PUBLISHED GENESIS → RUNNING
Verification npm run lps:verify scripts/merkle.ps1
Academic identity ORCID + DOI ORCID + DOI (shared)

Both protocols share:

  • The same author identity (ORCID: 0009-0008-8425-939X)
  • The same on-chain infrastructure (Polygon LiteraryAnchor)
  • The same Merkle tree methodology
  • The same provenance evidence chain structure

XI. License

  • Protocol specification (SOP-1): CC-BY-4.0
  • Reference implementation (Genesis Protocol): MIT
  • Research paper: CC-BY-4.0

"Systems that can prove their own integrity don't need trust."

Built with Rust. Verified with SHA-256. Anchored on Polygon.
DOI: 10.5281/zenodo.18729652 · ORCID: 0009-0008-8425-939X