Skip to content

Kubenew/SovereignStack

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

54 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

SovereignStack

SovereignStack

OASA Compatible L1 Sovereign-Ready L2 Secure-Runtime L3 Strict-Sovereign

44 RFCs (0001–0055) · 34 URI schemes · 18 protocols

🇬🇧 English · 🇩🇪 Deutsch · 🇫🇷 Français · 🇪🇸 Español · 🇮🇹 Italiano · 🇳🇱 Nederlands · 🇵🇱 Polski · 🇨🇿 Čeština · 🇸🇪 Svenska

🇯🇵 日本語 · 🇨🇳 中文 · 🇰🇷 한국어 · 🇮🇳 हिन्दी · 🇹🇷 Türkçe · 🇦🇪 العربية · 🇵🇹 Português · 🇷🇺 Русский · 🇺🇦 Українська

Community & Events

  • CFP submitted to Open Source Summit Europe 2026
  • Building Open Standards for Sovereign AI Systems: Provenance, Governance, and Interoperability

The Sovereign Intelligence Network — A distributed operating system for intelligence.

Architecture Diagram

SovereignStack is a protocol stack and runtime that treats intelligence itself as a networked resource. It is to autonomous intelligence what TCP/IP is to data networking.

1970s → ARPANET          (packet switching)
1990s → Internet         (global connectivity)
2000s → Cloud            (elastic compute)
2020s → AI Platforms     (model serving)
2030s → Sovereign Intelligence Networks (SovereignStack)

Design Principles

Every SovereignStack object is:

  1. Identifiable — has a unique URI
  2. Addressable — resolvable across the network
  3. Discoverable — findable via capability/semantic search
  4. Verifiable — cryptographically signed, with provenance
  5. Portable — movable across nodes and jurisdictions
  6. Federatable — shareable across sovereign boundaries
  7. Auditable — full history and reasoning trail

Universal Addressing

agent://researcher-1            # Agent identity
session://abc123                # Session
artifact://def456               # Produced artifact
memory://xyz789                 # Memory object
reason://decision-42            # Reasoning chain
knowledge://physics/newton      # Knowledge object
capability://legal-review       # Skill/capability
workflow://contract-analysis    # Workflow definition
contract://task-88              # Agent contract
org://acme                      # Organization
robot://drone-12                # Physical device
policy://gdpr-eu                # Governance policy
tel://zcube-a/evt-001           # Telemetry event (RFC-0033)
kv://zcube-a/gpu-003/session-a  # KV cache object (RFC-0034)
gateway://eu-frankfurt          # Federation gateway (RFC-0035)
mem://zcube-a/gpu-003/hbm       # Memory pool (RFC-0036)
profile://zcube-standard-v1     # Cluster profile (RFC-0037)
model://huggingface/Qwen/Qwen2.5-72B  # AI model (RFC-0040)
dataset://huggingface/c4        # Dataset (RFC-0040)
training://zcube-a/run-0042     # Training run (RFC-0040)
evaluation://zcube-a/eval-007   # Model evaluation (RFC-0040)
audit-pkg://node-001/2026-06-03 # Audit evidence package
continuity://zcube-a/legal-agent # AI continuity manifest (RFC-0052)
recovery://zcube-a/incident-42  # Recovery procedure (RFC-0050)
failover://zcube-a/evt-001      # Failover event log (RFC-0051)
playbook://zcube-a/gpu-failure  # Recovery playbook (RFC-0050)

Architecture

Protocol Stack
SovereignStack Protocol Stack — RFCs organized by layer

Deployment Models
Deployment Models: Air-gapped · Federated Mesh · Hybrid

Data Flow
OASA Data Flow — API Gateway → Inference → Audit Trail

ss-kernel
 ├── Identity        (UAI, SIG, trust, reputation)
 ├── Capabilities    (registry, discovery, routing)
 ├── Messaging       (event bus, streams)
 ├── Memory          (tiered: session → civilizational)
 ├── Scheduling      (compute placement, model selection)
 ├── Federation      (sovereign routing, replication)
 ├── Governance      (policies, jurisdiction, compliance)
 └── Provenance      (lineage, evidence, audit)

Core Features & Services

Alongside the core Rust primitives, SovereignStack implements a suite of Python-based microservices providing production-ready infrastructure:

  • OASA API Gateway: Secure, OpenAI-compatible entry point enforcing Data Loss Prevention (DLP), Strict Compliance Locking, and SPIFFE workload identity validation.
  • Federation Relay & Sync Engine: Decentralized node synchronization utilizing advanced Conflict-Free Replicated Data Types (CRDTs) to ensure eventual consistency across sovereign boundaries.
  • Merkle Audit Log: Cryptographically verifiable, append-only event stream providing a tamper-proof provenance trail for all system operations.
  • Predictive Scheduler: Autonomous operational controller employing exponential smoothing models to predict compute load and proactively scale resources.
  • Weight Federation: Secure registration and sharding of model weights across distributed nodes for collaborative inference.

Repository Structure

SovereignStack/
├── ss-kernel/        # Core kernel (identity, resolver, eventbus, registry, capability, policy)
├── ss-core/          # Shared types, URI parsing, errors
├── ss-crypto/        # Ed25519 signing, hashing
├── ss-identity/      # Universal Agent Identity
├── ss-capability/    # Capability declaration & matching
├── ss-eventbus/      # Event sourcing infrastructure
├── ss-cas/           # Content-addressed storage
├── ss-federation/    # Sovereign routing & discovery
├── ss-runtime/       # Multi-model execution runtime
├── ss-memory/        # Tiered memory subsystem (Stub)
├── ss-sessiond/      # Session lifecycle daemon
├── ss-scheduler/     # Compute placement (Stub)
├── ss-swarm/         # Multi-agent coordination (Stub)
├── ss-reason/        # Reasoning object store (Stub)
├── ss-kas/           # Knowledge addressing system (Stub)
├── ss-sig/           # Sovereign identity graph (Stub)
├── ss-trust/         # Trust framework (Stub)
├── ss-reputation/    # Reputation scoring (Stub)
├── ss-policy/        # Governance & jurisdiction (Stub)
├── ss-provenance/    # Computational lineage (Stub)
├── ss-jurisdiction/  # Jurisdiction compliance engine
├── ss-twin/          # Digital twin framework
├── ss-device/        # Reality interface layer (Stub)
├── ss-economy/       # Resource markets (Stub)
├── ss-sip/           # Sovereign Intelligence Protocol (Stub)
├── reference-node/   # Minimal reference node binary
├── conformance/      # Conformance test suites & profiles
├── rfcs/             # Protocol specifications (RFC-0001–0040)
├── docs/
│   ├── architecture/ # 10-page architecture breakdown
│   └── architecture/diagrams/ # SVG architecture diagrams
├── tests/            # Conformance test suite
├── examples/         # Reference implementations
└── playground/       # Try-it-now deployment

Standards Ecosystem

SovereignStack is governed by a formal standards framework:

Document Description
STANDARDS.md Root standards framework — 6 layers
OBJECT_MODEL.md Universal object model with cryptographic signatures
URI_STANDARD.md URI scheme registry & resolution rules
TRUST_MODEL.md Capability-based zero-trust security
SECURITY.md Threat model & incident response
CONFORMANCE.md 3-tier certification program & badges
CERTIFICATION.md Badge levels, colors, shapes, materials
PROTOCOL_REGISTRY.md 18 protocols with lifecycle management
REFERENCE_IMPLEMENTATIONS.md 13 core subsystems + 3 language bindings
ROADMAP-2035.md 10-year vision through Sovereign Intelligence Internet
BUILD_WINDOWS.md Windows build guide — exe, MSI, Python bundling
docs/i18n/README.md Internationalization — 18 languages with full i18n policy

Internationalization

SovereignStack is translated into 18 languages. The project follows the Debian model: English is canonical, translations are maintained copies. See the i18n index for available languages and the translation policy.

RFCs

RFC Title Status
RFC-0001 Sovereign Object Model Draft
RFC-0002 URI Standard + ABNF Grammar Draft
RFC-0003 Trust Graph Draft
RFC-0004 Capability Registry Draft
RFC-0005 Knowledge Objects Draft
RFC-0006 Reasoning Objects Draft
RFC-0007 Event Bus Draft
RFC-0008 Federation Routing Draft
RFC-0009 Session Lifecycle Draft
RFC-0010 Conformance Framework Draft
RFC-0011 Identity & DID Resolution Draft
RFC-0012 Cryptographic Signatures & Verification Draft
RFC-0013 Provenance Graph Draft
RFC-0014 Agent Communication Protocol Draft
RFC-0015 Workflow Execution Engine Draft
RFC-0016 Resource Allocation & Accounting Draft
RFC-0017 Sovereign Memory Protocol (Detailed) Draft
RFC-0018 Content-Addressable Storage Draft
RFC-0019 Trust & Reputation Scoring Draft
RFC-0020 Policy Enforcement Draft
RFC-0021 Jurisdiction & Data Residency Draft
RFC-0022 Audit Log Streaming Draft
RFC-0023 Extension & Plugin System Draft
RFC-0024 Multi-Model Routing Draft
RFC-0025 Session Migration Draft
RFC-0026 Capability Delegation Chain Draft
RFC-0027 Swarm Coordination Draft
RFC-0028 Digital Twin Synchronization Draft
RFC-0029 Intelligence Economy Primitives Draft
RFC-0030 Network Topology Awareness Draft
RFC-0031 KV Locality Scheduling Draft
RFC-0032 AI Fabric Protocol Draft
RFC-0033 Fabric Telemetry Protocol Draft
RFC-0034 Distributed KV Placement Draft
RFC-0035 Topology-Aware Federation Draft
RFC-0036 Memory Fabric Objects Draft
RFC-0037 AI Cluster Profiles Draft
RFC-0040 Model Lineage Protocol Draft
RFC-0050 AI Continuity & Disaster Recovery Draft
RFC-0051 Model Failover Protocol Draft
RFC-0052 AI Continuity Manifest Draft
RFC-0053 Sovereign Recovery Profiles Draft
RFC-0054 Compliance Framework (OASA CCM) Draft
RFC-0055 Windows Build System Draft

Quickstart

# Clone & enter
git clone https://github.com/Kubenew/SovereignStack.git
cd SovereignStack

# Windows: build .exe binaries (see BUILD_WINDOWS.md for full MSI guide)
cargo build --release --bin ss-node --bin ss-cli

# Launch playground (Docker Compose)
docker compose -f playground/docker-compose.yml up -d

# Or build from source
cargo build --workspace

Test

cargo test --workspace

Compliance & Certifications

SovereignStack provides a complete audit and certification framework for enterprise AI deployments:

Asset Description
OASA CCM Core Controls Matrix — 18 controls across NET/RUN/HW/AUD/CONT for L1/L2/L3
SOC 2 Mapping Full mapping to AICPA TSC 2023 — all 5 trust categories
Gaia-X Mapping Mapping to Gaia-X Trust Framework + Self-Description generation
SOC 2 Type 1 Report Template Pre-built evidence package template for auditor review
Compliance Schema JSON Schema (Draft 2020-12) for automated node validation
Audit Evidence Schema Structured evidence package for oasa-audit report
Validation Script Automated compliance scanner with --audit-host

Target frameworks: ISO 42001, EU AI Act, SOC 2, NIS2, Gaia-X, CSA CCM, FedRAMP

# Generate compliance evidence package
oasa-audit report --framework iso42001 --framework soc2

# Export for auditor review
oasa-audit export-pdf --output audit-report-2026-Q2.pdf

License

Apache-2.0 OR MIT (dual-licensed)

About

**Open Architecture Specification for Autonomous and Sovereign AI (OASA)**

Resources

License

Code of conduct

Contributing

Security policy

Stars

Watchers

Forks

Packages

 
 
 

Contributors