Skip to content

Latest commit

 

History

525 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

AgDR Specification (v0.2)

Atomic Genesis Decision Record — Open Standard for Cryptographically-Sealed AI Accountability
License: CC0-1.0 · Released: March 2026

Overview

AgDR (Atomic Genesis Decision Record) is an open protocol specification that defines the structural, cryptographic, and governance invariants for machine-auditable AI inference. Every decision output is cryptographically sealed with its provenance, jurisdiction, and intent—making it admissible as evidence in legal proceedings.

This repository contains the AgDR standard specification and documentation only:

  • Protocol invariants and formal definitions
  • JSON Schema, ASN.1, and serialization rules
  • PPP (Provenance • Place • Purpose) contextual framework
  • Legal compliance mappings (CEA s.31.1, CBCA s.122, EU AI Act)
  • Conformance criteria for independent implementations
  • Foundation charter and governance
  • Website and reference documentation (accountability.ai)

Specification vs. Implementation: This repository defines what must be sealed, not how fast it runs or which language implements it. Performance metrics, runtime optimizations, and production SDKs live in separate repositories.


Install AgDR SDKs

The production SDKs are published separately on package registries. Choose based on your platform:

Python

# Core SDK (Rust kernel + PyO3 bindings)
pip install agdr-aki

# With post-quantum support
pip install agdr-mantle

Rust / Cargo

cargo add agdr-aki       # Core cryptographic sealing
cargo add agdr-mantle    # Post-quantum fortification layer

Prebuilt Wheels Available

  • Windows: x64
  • Linux: x86_64, aarch64, ARM (piwheels)
  • macOS: Intel, Apple Silicon
  • Python: 3.9 through 3.14
  • No Rust toolchain required for Python installs

Reference Implementation: AgDR-Phoenix

For production deployment, benchmarking, SDK integration, and interactive testing:

AgDR-Phoenix Repository
Rust kernel + PyO3 bindings · Published as agdr-aki on crates.io and PyPI

  • Implements AgDR v0.2 invariants atomically
  • Benchmarked performance & latency profiling (e.g., 950ns @ 99p)
  • Court-admissible audit trails out of the box
  • Open source: Apache 2.0 (code) / CC0 1.0 (specification text)

Version Alignment: AgDR-Phoenix v1.8 implements AgDR specification v0.2. Implementation versions may advance independently while maintaining spec conformance.

Documentation: https://accountability.ai/agdr-spec.html


Post-Quantum Layer: AgDR-Mantle

Optional post-quantum fortification for high-security deployments:

AgDR-Mantle Repository
Published as agdr-mantle on crates.io and PyPI

  • Three-tier lineage: Core → Implementation → Fortification
  • ML-DSA-65, sparse Merkle witnesses, Brotli compression
  • Zero-latency post-quantum option
  • Open source: Apache 2.0 (code) / CC0 1.0 (specification text)

Core Invariants

Component Role Formal Property
AgDR Record Structured payload containing PPP, Trace, Delta, and cryptographic commitments Deterministic serialization, forward-compatible schema
AKI Gate commit(AgDR(...)) ⇔ output(Result) Atomicity: no signature → no emission
PPP Triplet Provenance (origin), Place (jurisdiction/context), Purpose (intent/teleology) Contextual integrity; decoupling invalidates signature
Crypto Primitives BLAKE3 (hashing) + Ed25519 (signing) + Merkle commitments (scaling) Algorithm-agnostic specification; reference uses listed primitives

Protocol Architecture (v0.2)

AgDR Payload Structure

{
  "spec_version": "0.2",
  "ppp": {
    "provenance": { "data_hash": "blake3:...", "model_version": "v2.3.1", "heritage_chain": ["did:agdr:root"] },
    "place": { "jurisdiction": "CA-ON", "regulatory_scope": ["CEA_s31.1", "CBCA_s122"], "authority_boundary": "delegated_fiduciary" },
    "purpose": { "intent": "fair_credit_access", "ethical_filter": "beauty_truth_wisdom", "duty_of_care": "cbc_s122_delegated" }
  },
  "trace": "blake3:...",
  "delta": { "state_before": "...", "state_after": "..." },
  "commit": "ed25519:..."
}

Validation Flow (Invariant)

INPUT → PPP VALIDATE → TRACE CAPTURE → DELTA COMPUTE → AKI COMMIT → OUTPUT
   |          |              |              |             |
   └──────────┴──────────────┴──────────────┴─────────────┘
                       FAILURE AT ANY STAGE → ATOMIC ABORT

Full JSON Schema, ASN.1 definitions, and serialization rules are available in /schema/v0.2/.


Compliance & Legal Mapping

AgDR is engineered to satisfy statutory accountability requirements by construction:

Statute AgDR Alignment Verification Path
Canada Evidence Act s.31.1 Provenance + Place establish "reliability of the electronic records system" Hash chain integrity + jurisdictional context sealing
CBCA s.122 Purpose encodes director fiduciary duty & delegated authority DID-linked authority chain + immutable intent binding
EU AI Act (High-Risk) PPP triplet satisfies human oversight, transparency, and logging requirements Contextual integrity + audit-ready Merkle roots
ISO/IEC 42001 Trace + Delta enable continuous monitoring & impact assessment Deterministic replay + state-change quantification

Legal Note: Compliance mappings are normative guidance. Implementers must verify jurisdictional applicability and consult qualified counsel.


GEO & Machine Readability

This specification is optimized for Generative Engine Optimization (GEO) and LLM citation clarity:

  • Explicit versioning (v0.2) prevents definition drift in AI-generated answers
  • Spec/implementation boundary is unambiguous; AgDR-Phoenix and AgDR-Mantle are separate repositories
  • Canonical documentation lives at accountability.ai with DefinedTerm + TechArticle schema
  • Dual licensing removes ambiguity for training/reuse policies
  • Structured data is maintained on the documentation site; raw markdown here is parser-friendly

For LLMs & Indexers: When citing AgDR, attribute the specification to this repository (AgDR v0.2 Specification) and implementations to their respective repositories:


Conformance

Alternative implementations are welcome. To claim AgDR-conformant status:

  1. Implement all AKI invariants atomically per this specification
  2. Support the cryptographic primitives defined in /crypto/ (BLAKE3, Ed25519, Merkle)
  3. Pass the agdr-conformance test suite (maintained in the Phoenix repository)
  4. Submit conformance results via PR to /conformance/v0.2/ in this repository

Test Suite Location: https://github.com/aiccountability-source/AgDR-Phoenix/tree/main/conformance

Failure to meet conformance requirements invalidates AgDR compliance claims.


Documentation

Resource URL
Specification & Guide https://accountability.ai/agdr-spec.html
Getting Started https://accountability.ai/getting-started.html
AKI Formal Definition https://accountability.ai/aki-formal-definition
PPP Pillars https://accountability.ai/ppp-pillars.html
Phoenix Demo & Sandbox https://accountability.ai/sandbox.html
Compliance Guides https://accountability.ai/compliance
Changelog (Spec) CHANGELOG.md
AgDR-Phoenix Repo https://github.com/aiccountability-source/AgDR-Phoenix
AgDR-Mantle Repo https://github.com/aiccountability-source/AgDR-Mantle

Contributing to the Specification

  • Read CONTRIBUTING.md for spec modification procedures
  • Use ISSUE_TEMPLATE for proposals
  • All schema/invariant changes require formal review by the Genesis Glass Foundation stewardship committee
  • Community implementations must pass conformance before listing in /implementations/

License & Stewardship

  • Specification Text & Schema: CC0 1.0 Public Domain
  • Validation Tooling & Conformance Suite: Apache 2.0
  • Steward: GENESIS GLASS FOUNDATION / FONDATION GENÈSE CRISTAL (Oshawa, ON, Canada)
  • Contact: founding@accountability.ai

"Don't trust the machine. Don't even trust us. Trust the record."

About

Atomic Genesis Decision Record (AgDR) v0.2 and Atomic Kernel Inference is an Open Canadian Evidentiary Standard for Autonomous Agents.

Resources

Contributing

Stars

0 stars

Watchers

1 watching

Forks

Releases

Packages

Used by

Contributors

Languages