Skip to main content
TRUST MODEL

Trust and Scope

This page describes exactly what Cryptographic Runtime Governance evidence bundles prove, what they do not prove, and the trust assumptions underlying the system. Read this like a threat model, not marketing.

What Evidence Bundles Prove

A specific tool call decision was recorded

Each receipt contains the tool name, decision (PERMITTED or DENIED), request ID, and timestamp. The receipt is signed with Ed25519 over the canonical JSON representation.

The signer held a specific private key at signing time

Ed25519 signatures are deterministic. A valid signature proves the signer possessed the private key corresponding to the embedded public key.

The receipt chain is tamper-evident

Each receipt includes the SHA-256 hash of the previous receipt (including its signature). Any modification to a receipt breaks the chain from that point forward.

The Merkle tree binds all receipts to a single root

Inclusion proofs allow any single receipt to be verified against the bundle's Merkle root without access to other receipts.

Verification works offline and air-gapped

All cryptographic operations use Ed25519 and SHA-256. No network access, certificate authorities, or external services are required for verification.

What Evidence Bundles Do NOT Prove

That the tool call actually executed

A receipt records a governance decision, not an execution trace. A PERMITTED receipt means the gateway approved the call. It does not prove the tool ran, returned a specific result, or completed successfully.

That the AI agent was actually constrained

The receipt chain proves decisions were recorded and signed. It does not prove the agent could not bypass the gateway. Enforcement is a separate concern addressed by a private runtime component.

That the policy was correct or complete

The bundle references a policy hash, not the policy itself. A bundle with all PERMITTED decisions may reflect a permissive policy, not a secure one.

That the timestamps are accurate

Timestamps are self-reported by the signing gateway. Roughtime IETF corroboration is available but optional. Without Roughtime enabled, a compromised gateway could backdate or future-date receipts.

That the public key belongs to a specific entity

Evidence bundles embed a raw Ed25519 public key. Self-signed X.509 identity binding via SPIFFE is available but optional. Without identity binding enabled, the key proves cryptographic consistency, not organizational identity.

Trust Assumptions

The signing key is not compromised

If an attacker obtains the Ed25519 private key, they can forge receipts indistinguishable from legitimate ones. Key management is the deployer's responsibility.

The gateway is positioned to observe tool calls

The MCP gateway must be in the request path between the AI agent and tool providers. Calls that bypass the gateway are not recorded.

The verifier has the correct public key

Verification requires the verifier to trust that the public key in the bundle corresponds to the expected signer. Out-of-band key distribution is assumed.

Cryptographic primitives remain secure

The system relies on SHA-256 collision resistance and Ed25519 unforgeability. If these primitives are broken, all guarantees are void.

Enforcement Boundary

The public components described on this site (MCP gateway, Python SDK, evidence bundles) provide observation and proof. A separate, private runtime component provides enforcement by ensuring the agent cannot bypass the governance gateway. The enforcement runtime is not open-source and is available under separate licensing. Contact us for details.

ASSURANCE TIERS

Deployment Assurance Levels

AGA supports three tiers of deployment assurance. Each tier adds cryptographic and infrastructure guarantees that reduce the trust surface.

TierComponentsTrust AssumptionUse Case
BronzeSealed artifact + signed receiptsPortal operator is honestDevelopment, testing
Silver+ Continuity chain + external timestampsk-of-n operators not all colludingProduction, enterprise
Gold+ HSM key isolation + TEE attestation + external anchorHardware enclaves are uncompromisedDefense, critical infrastructure
IRREDUCIBLE RESIDUALS

What Cannot Be Eliminated

After every engineering mitigation is applied, three risks remain. Every system terminates in assumptions. Ours are smaller than any competing approach, and we name them explicitly.

Total Collusion

If all k-of-n TEE-hosted Portal operators collude simultaneously, receipts can be forged. This is the terminal trust assumption of any distributed cryptographic system.

Policy Quality

AGA proves a policy was authored, approved, enforced, and measured. It cannot prove the policy was correct for the situation. Policy quality is a human judgment, intentionally out of scope.

Legal Precedent

Cryptographic governance evidence has no case law yet. The first legal proceeding that accepts it establishes precedent. Everything before that is engineering preparation.

POST-QUANTUM

Quantum-Resistant Signing

AGA supports ML-DSA-65 + Ed25519 hybrid composite signatures (NIST FIPS 204). Both signatures are produced over the same message and both must verify for acceptance. No downgrade attacks are possible.

This provides quantum resistance today while maintaining compatibility with existing Ed25519 infrastructure. If Shor-class quantum computers break Ed25519 (per Babbush et al., Google Quantum AI, March 2026), ML-DSA-65 remains valid. If ML-DSA-65 has an undiscovered classical weakness, Ed25519 remains valid. An attacker must break both simultaneously.