How to implement agent routing and autonomy: intent dispatch at ingress, orchestration in the loop, versioned manifests with pinned tools, four autonomy patterns, shared contracts, and control points. Model selection lives in the LLM Blueprint. MCP is transport only.
Two views, one G.A.I.N subject: estate operating model and PGAR on the request path.
Estate reference design for G.A.I.N Governance: inventory, obligation map, and evidence pack before scale. Canonical model lives on G.A.I.N Governance. Runtime is PEP/PDP.
How to implement a Policy-Governed Agent Runtime. PEP/PDP enforcement, subject-action-resource-context contracts, audit replay, step-up, and boundary-specific playbooks.
Reference design for G.A.I.N MCP as transport: client, gateway, server registry. Tool contracts (manifests) live in the Agents Blueprint.
Reference design for G.A.I.N RAG: route-declared retrieval, PEP-gated search, context packs, and validation before synthesis.