Skip to main content

Agents Playbooks

Playbooks · Agents overview · Intent router (Plane ①) →

Implementation guides for the Agents Blueprint. These playbooks are the how for intent dispatch and the agent loop. Model selection is a separate series: LLM.

THE CLAIM

Route before the loop. Plane ① picks the governed path; Plane ② runs the agent loop. Each plane has its own contracts, eval gates, and ownership. The LLM gateway is G.A.I.N LLM, not this series.

Two planes, plus manifests

GroupOverviewWhat you build
Intent routerOpen →Route contract, lifecycle, layered classifier, agentic app wiring, routing eval CI
OrchestrationOpen →Session custody, memory, autonomy shape, inference handoff; PEP in Runtime
ManifestsOpen →Tool contract, pdp_action, version, pin. Transport-agnostic (MCP is one adapter)

Model routing (capability matrix, gateway, canary) lives in LLM playbooks.


  1. Intent router: contract reference → lifecycle → eval CI in parallel → classifier → wire app
  2. Manifests: registry shape, then lifecycle (the route pins tool_manifest)
  3. Orchestration: copy the route freeze into the run pin → autonomy pattern → every inference through the gateway, then Runtime for PEP
  4. LLM when you have multiple models or regions: capability matrix → task routing → canary

Bridge reading: PGAR Agentic app · Agents Blueprint · G.A.I.N LLM · Eval Plane ①: Input.

Intent router playbooks at a glance

PlaybookOne-line purpose
Route contract referenceField dictionary: route, activation_target, manifest, policy, retrieval, memory
Route table lifecycleVersioned route contracts, storage patterns, promote and rollback
Layered classifierEligible routes, rules (channel / event), classifier, LLM fallback, safety
Wire agentic appFront door, decide-only handoff, event bind, PGAR start
Routing eval CIGolden set, release gates, incident replay

Orchestration playbooks at a glance

PlaybookOne-line purpose
Session custodyDurable run pin, token, loop bounds; credentials never reach the model
MemoryIsolation, TTL, retrieve_only prefs and episodes
Autonomy shapePattern 0-3 from the route: one call, loop, fixed stages, or hybrid
Inference handoffLLM gateway for every LLM call; manifest check then PEP

Enforcement detail stays in Runtime (foundation, boundary, assurance). Pattern contracts: Agents Blueprint.

Manifests at a glance

PlaybookOne-line purpose
Manifest registryTool schema, pdp_action, risk tier; unknown tools never reach PEP
Manifest lifecycleWhere manifests live, version, pin, promote, roll back

MCP is one way to expose these tools. See MCP.

Who should read what

Intent router → · Orchestration · Manifests · LLM · Enterprise AI Workflow Patterns (choose Pattern 0-3)