Compliance

SovereignClaw connects AI agent runtime security to evidence-ready compliance controls for healthcare, finance, government, and other regulated environments.

SovereignClaw maps its controls to SOC 2, NIST, HIPAA, ISO 27001, and OWASP Agentic (AA1 to AA10). Every permitted execution produces a signed receipt in an append-only, hash-chained ledger with Merkle anchors, providing the audit evidence that regulated industries need. These are control mappings, not certifications.

Compliance on this platform starts at the execution layer. Every permitted action generates a signed Authority Receipt anchored in an append-only Merkle ledger, which gives teams evidence they can route into audit workflows, SIEM tools, and policy reviews. To see how the receipts are generated, review the execution pipeline; to understand the guarantees behind them, read the formal security properties. Published on SSRN (ID 6290760). The DetGate execution-gating method behind SovereignClaw is patent pending with the USPTO.

Healthcare Compliance

PHI access governance and clinical workflow enforcement. Evidence for California AB 489 programs through cryptographic execution gating. Every access to patient data is tier-classified, policy-evaluated, and receipt-verified. HIPAA-aligned audit trails with immutable Merkle-anchored receipts.

Financial Compliance

Fiduciary obligation enforcement for autonomous financial workflows. Deterministic execution controls for governed financial actions. Cryptographic audit trails for every trade, reimbursement, and transfer. Threshold signature requirements for elevated financial operations.

Government & DOD

Support for DOD AI-First operations. Air-gapped deployment options for classified environments. NIST control mappings that support FedRAMP-aligned programs; SovereignClaw is not FedRAMP authorized. On-premise deployment with full compliance metrics export.

Compliance Controls

EU AI Act

For agentic AI systems in scope of the EU AI Act, SovereignClaw helps operationalize the high-risk control areas - risk management, record-keeping, human oversight, accuracy, robustness, and cybersecurity - by enforcing runtime authorization and emitting verifiable execution evidence. It does not replace compliance work; it provides the runtime control and evidence that make agentic AI governable. See EU AI Act compliance for AI agents for the full control-area mapping, plus deep dives on high-risk AI systems, human oversight, logging and record-keeping, and cybersecurity controls.

Audit Trail Architecture

Every permitted execution emits an Authority Receipt dual-signed with Ed25519 and post-quantum SLH-DSA-SHA2-256f, chained in an append-only ledger with Merkle anchors (Security Property S8). The full receipt chain is externally verifiable against operator-pinned public keys without private key access. Compliance metrics are exported in both JSON and CEF format for SIEM integration. Teams planning rollout can compare these controls against the deployment tiers and implementation model.

Request Early Access

Frequently Asked Questions

Does SovereignClaw support SOC 2 compliance?
Yes. SovereignClaw maps its controls to SOC 2 criteria, with receipt chain evidence and policy bundle exports for auditors. Every permitted execution generates a signed Authority Receipt in an append-only, hash-chained ledger with Merkle anchors. These are control mappings, not a certification or audit report.
Does SovereignClaw support FedRAMP?
SovereignClaw is not FedRAMP authorized. Its controls map to NIST, which supports FedRAMP-aligned programs, and air-gapped and on-premise deployment options suit government and DOD workloads. Authorization is granted to a specific system through an agency assessment.
How does SovereignClaw address OWASP Agentic Top 10?
SovereignClaw maps its controls to OWASP Agentic risks AA1 to AA10, addressing prompt injection, unauthorized tool use, excessive permissions, and other agentic security risks through deterministic enforcement.
Which compliance frameworks does SovereignClaw map to?
SovereignClaw maps its controls to SOC 2, NIST, HIPAA, ISO 27001, and OWASP Agentic (AA1 to AA10). Every permitted agent action produces a signed receipt in an append-only, hash-chained ledger, providing audit evidence for reviewers. These mappings support your compliance program; they are not certifications or audited compliance claims.