Skip to main content
AGAAttested Governance Artifacts

Cryptographic Runtime Governance

For Agentic AI

Enforce behavior at execution.
Prove every decision cryptographically.
Verify anywhere, offline and air-gapped.

Policy without proof
is just a suggestion.

See How It Works What is Cryptographic Runtime Governance?

Architecture

Seal. Enforce. Prove.

01

Seal

Policy Artifact

Ed25519 + ML-DSA-65 signatures · SHA-256 integrity

Bind approved behavior to a signed, immutable policy object at build time. The sealed hash is the single source of truth for what the agent is authorized to do.

02

Enforce

Portal Runtime Boundary

Launch Gate · Drift Detection

The Portal validates policy before execution, monitors continuously, and executes enforcement actions autonomously. The agent holds no keys and cannot self-authorize.

Private runtime component

03

Prove

Evidence Bundle

Merkle proofs · Hash-linked receipts

Export a portable, tamper-evident proof package. Every receipt is signed and hash-linked. Any third party can verify offline - 8-step verification with independent decision re-derivation.

Standard cryptographic primitives. No special hardware. Fully offline-capable. See the full walkthrough

Agentic AI Governance at the Execution Boundary

Autonomous agents create action risk: unauthorized tool invocations, state modifications, and financial transactions that content filters cannot address. Current approaches stop at observation. None produce cryptographic proof of enforcement.
Learn more about agentic AI governance.

Autonomous Action Without Confirmation

Agentic AI systems invoke tools, modify state, and execute transactions without human approval at each step. Governance must operate at the speed of execution, not after the fact.

Frameworks Define Intent

Every major governance framework calls for enforceable runtime controls. Frameworks define what should happen. Proving what actually happened requires cryptographic enforcement at the execution boundary.

Monitoring Watches After Execution

Log aggregation and anomaly detection watch behavior after execution. Cryptographic proof of enforcement requires a mandatory runtime boundary operating at the speed of execution.

AGA Closes the Gap

Attested Governance Artifacts seal policy before execution, enforce at a mandatory runtime boundary where the agent holds no keys, and produce signed receipts that verify offline.

Differentiation

Where Existing Systems Stop

Existing tools cover fragments of the governance problem. AGA closes the gaps they leave open.

Logging / SIEM

Provides: Event capture, anomaly alerts

Gap: Logs can be altered or deleted; no enforcement

Policy-as-Code

Provides: Declarative rules, drift alerts

Gap: Advisory only; no runtime binding

TEE / SGX

Provides: Hardware-isolated execution

Gap: No offline verification; vendor-locked

Blockchain Audit

Provides: Immutable ledger, timestamping

Gap: No runtime enforcement; latency constraints

AGA (this system)

Provides: Sealed policy, runtime enforcement, portable cryptographic proof

Gap: Extends all layers above

What AGA Adds

Sealed Reference Hashes

An immutable, signed snapshot of authorized behavior. A cryptographic commitment that cannot be revised after the fact.

Mandatory Runtime Boundary

The Portal enforces policy before execution begins. The subject holds no signing keys and cannot self-authorize.

Signed Enforcement Receipts

Every enforcement action produces a signed receipt, hash-linked into an append-only chain. Each receipt is independently verifiable.

Offline-Verifiable Evidence

Evidence bundles contain everything needed to verify compliance. No network, no callback, no trust in the producing system.

AGA extends attestation with runtime enforcement and portable cryptographic proof.

What Becomes Possible

  • Prove exactly which policy was enforced, when, and by whom

  • Verify compliance offline in air-gapped or contested environments

  • Detect drift between authorized and actual behavior at runtime

  • Export tamper-evident evidence that any third party can audit

What Becomes Impossible

  • A sealed policy artifact cannot be modified without invalidating its cryptographic signature

  • A valid receipt cannot be produced without the Portal's signing key

  • A receipt cannot be deleted or reordered without breaking the hash-linked chain

  • A tampered evidence bundle cannot pass independent cryptographic verification

Market Opportunity

Built for Environments Where
Logging Is Not Enough

Defense & DoD

Autonomous systems, drones, agents

Sealed policy artifacts and cryptographic chain of custody from build to deployment to mission execution. Verifiable in disconnected and contested environments.

  • Autonomous fleets
  • Contested ops
  • Air-gapped systems

SCADA / ICS

Critical infrastructure

Tamper-evident, cryptographic audit trails for industrial control systems. Deterministic resource bounds and offline-verifiable evidence bundles designed for real-time OT environments.

  • Energy grids
  • Water & utilities
  • Pipeline control

Regulated

Finance, healthcare, legal

Governance enforcement where regulatory compliance requires cryptographic proof. Immutable audit trails built for examiner scrutiny.

  • Trading oversight
  • HIPAA AI systems
  • EU AI Act

Enterprise AI

Agentic AI governance at scale

Cryptographic proof that only authorized models ran within enforced operational parameters. Immutable, independently portable audit trails that no one can alter or silently delete.

  • Multi-agent ops
  • Model governance
  • MCP enforcement
Critical Vulnerability

AI Systems Are
Operating Without Proof

Every day, critical infrastructure runs on AI systems with no cryptographic guarantee they have not been compromised. Compliance teams demand verifiable evidence. Current solutions provide neither.

Model Poisoning

Adversaries poison models during training, fine-tuning, or deployment. No cryptographic proof of origin means no way to verify what is actually running in production.

Silent Drift

Models change behavior post-deployment through prompt injection, weight modification, or gradient attacks. Current monitoring cannot detect subtle manipulation until it is too late.

Forensic Evidence Gap

When incidents occur, there is no audit trail that survives scrutiny. Investigations rely on logs that can be tampered with. Auditors need tamper-evident evidence.

Paper Governance

Contracts define what AI should do. Nothing enforces it at runtime. Legal language cannot govern machine behavior. Enforcement requires cryptographic constraints, not contractual ones.

Standards Alignment

Architecture designed for environments requiring:

Architectural design alignment. Not certification or compliance advice.

NIST
AI RMF 1.0
CISA
Secure by Design
NSA
AI/ML Supply Chain
EU AI Act
High-Risk Systems
CoSAI
MCP Security Analysis
SLSA
Supply Chain Integrity
SSDF
NIST SP 800-218
NIST
SP 800-207 (Zero Trust)
Core Properties

Active Enforcement Architecture

Active Enforcement

Mandatory runtime boundary.

The agent holds no signing keys and cannot self-authorize. Static policy blocks unauthorized tools. Behavioral detection catches pattern shifts. Phantom execution captures bypass attempts.

Standard Cryptography

No proprietary protocols.

Ed25519, ML-DSA-65, SHA-256, BLAKE2b-256, Merkle trees. Post-quantum hybrid signatures available. Auditable by anyone, anywhere, offline.

Offline Verification

No callback required.

Evidence bundles are fully self-contained. 8-step verification with independent decision re-derivation. Verify in air-gapped, classified, or contested environments.

Independent Proof

Third-party verifiable.

The entity being governed cannot produce independent proof of its own governance. The rules, the enforcement, and the audit must be cryptographically independent.

Local-First

No cloud dependency

Real-Time Safe

Constant-time ops

Privacy-First

No payload disclosure

Provider-Agnostic

Any model or platform

MCP-Ready

Native MCP integration

Tamper-Evident

Hash-linked chain

Implementation Path

From Evaluation to Integration

01

Evaluate

Download the standalone demo. Run the independent verifier. Inspect the evidence bundle and receipt chains.

See the Walkthrough
02

Explore

Read the protocol specification. Review the reference implementation. 1,227 automated tests. Full documentation.

View Technology
03

Integrate

Language-agnostic protocol with a TypeScript/Node.js reference. Build in whatever stack you use.

Read the Specification
04

Collaborate

Interested in deploying AGA for defense, critical infrastructure, or enterprise AI governance? Let's talk.

Request Info
See the complete technical evaluation path