Architecture
The SovereignClaw architecture is built for AI agent runtime security: deterministic execution gating, 33 Rust crates, 1,345 workspace tests passed in the October 7, 2026 release run, and 9 formal security properties.
SovereignClaw's architecture treats every LLM-proposed action as untrusted input. The seven-stage execution pipeline turns model output into a deterministic, policy-evaluated decision before any side effect can occur. For teams evaluating enterprise AI governance, this page shows how runtime security, evidence generation, and operational control fit together in one system.
This architecture is designed for organizations that need more than best-effort AI guardrails. Instead of filtering or post-hoc review, SovereignClaw canonicalizes intent, verifies facts, assigns risk, and binds any permitted execution to signed artifacts and receipts. Pair this architecture view with the security properties, compliance coverage, and deployment tiers to understand the full operating model. Published on SSRN (ID 6290760). The DetGate execution-gating method behind SovereignClaw is patent pending with the USPTO.
Execution Pipeline
Every agent action passes through a deterministic 7-stage pipeline before any execution can occur. No shortcuts. No bypasses.
Stage-by-stage execution flow
Risk Tier Classification
Every action is classified into one of four risk tiers. The gate behavior is determined by the tier, not by the LLM's self-assessment.
Policy behavior by tier
| Tier | Name | Example | Gate Behavior |
|---|---|---|---|
| T0 | Observe | ls -la | Allowed by default policy |
| T1 | Standard | Write app config | Allow with signed receipt |
| T2 | Elevated | Restart production service | Requires threshold signatures (default quorum 1 of 2) |
| T3 | Sovereign | Delete IAM / DB / infra | Mechanical refusal: refused by default policy |
Crate Architecture (33 Crates)
The workspace is grouped by role: the kernel, platform services, skills and supply chain, agent and LLM integration, self-improvement, CLI tooling, binaries, and test crates. The October 7, 2026 release run passed 1,345 workspace tests with zero failures. That separation is what lets SovereignClaw map execution controls cleanly into audit and compliance evidence.
Core, control-plane, and runtime crates
| Crate | Role |
|---|---|
| libs/core | Immutable SovereignIR schema, canonical hashing, execution graph |
| libs/gate | Policy engine: fact recalculation, tier classification, threshold signatures, WASM plugin sandbox, revocation check |
| libs/adapter | Attested runtime: 14-step verification, nonce replay protection, AEAD credential store, SQLite revocation registry |
| libs/receipt | Append-only, hash-chained ledger of dual-signed authority receipts with Merkle anchors |
| libs/tenant | Multi-tenant isolation: context hashing, quotas, cross-tenant denial |
| libs/vault | Per-tenant, per-environment credential vaults with lease-gated access |
| libs/crypto | Ed25519 and SLH-DSA-SHA2-256f (FIPS 205) receipt signatures, threshold signature collection, and key rotation |
| libs/identity | SSO/SCIM identity bridge: IdP group → approval quorum mapping |
| libs/approval | Persistent async approval workflows with quorum, TTL, escalation |
| libs/telemetry | Observability and metrics export |
| libs/enterprise | 16 governed enterprise adapters (Terraform/Pulumi, kubectl, Vault/AWS Secrets Manager, psql, AWS CLI, gh, PagerDuty, ServiceNow/Jira) with tier classification (T0 to T3) |
| libs/skill_registry | Verified manifests, publication state, quarantine, intake freeze, tenant allowlisting |
| libs/skill_scanner | WASM bytecode behavioral analysis - flags network, file-write, env-read imports against declared scope |
| libs/wasm_runtime, libs/tools | Sandboxed WASM skill runtime and governed tool definitions |
| libs/mesh | Ed25519-signed inter-agent envelopes, delegation chains with scope attenuation, propagation depth limits |
| libs/agent | LLM orchestration loop, provenance-tagged memory, context injection boundary, swarm execution runtime |
| libs/provider, libs/memory | LLM provider integration and agent memory |
| libs/browser-adapter | Governed Playwright browser adapter, driving a separate stdio-only worker with no network listener |
| libs/strategy, libs/evaluator, libs/learning | Zero-authority Stage 1 learning: captures and drafts lessons into a tenant-local shadow store; cannot change memory, policy, Gate, or adapters |
| libs/cli-core, libs/cli-generator | CLI tooling shared by the binaries |
| libs/gateway | HTTP control plane with deployment attestation, compliance metrics (CEF/SIEM), production mode profile, signed POST /api/v1/evaluate, and governed A2A v1 |
| bin/sovclaw, bin/sovclaw-gateway, bin/sovclaw-tui | CLI, gateway server, and terminal UI |
| bin/sovclaw-improve, bin/sovclaw-browser-worker | Self-improvement runner and governed browser worker |
| tests/adversarial, tests/proptest | Adversarial and property-based test suites |