Skip to content
runward
runward
Product
Who it is for
Trust
Documentation
News
Try it on an example Talk to us

This appendix grounds the principles set out here in real artifacts. They live in three systems built separately, with no code sharing: a chain of test agents born of several experiments (system 1), a project-support platform built on a knowledge graph (system 2), and a personal assistant with long-term memory (system 3). The first two cover the full set of pillars and are set side by side below; the third, specialized, exercises the memory and evaluation pillars. That these three systems converge on the same patterns, without sharing a single line of code, is a serious clue. Let us stay precise about its scope: it is validation by use, not proof of independence, since the same hand designed them. Absence of shared code does not equal independence of design.

The artifacts below are described by their function, not by their location. What matters is not the name of a file but the fact that the block exists and that the systems converge on it, verifiable by a team with access to the repositories. No path, file name, or commercial name is cited.

Foundation, boundaries, and providers

PrincipleSystem 1, test agentsSystem 2, project platform
Pure domain, ports and adaptersPure domain and use cases, secondary adapters separated from the coreDomain ports isolated, dedicated persistence adapters
Explicit dependency injectionDependencies passed by constructor, centralized factoriesDependency injection container with a single entry point
Model boundary, tiered portModel provider port, three tiers fast, balanced, deepModel adapters behind a complexity router
Routing by model tierTier declared per agent, escalation in case of doubtTask classified as simple, moderate, complex
Feature detection, bimodalDetection at startup, minimal, full, integrated modesPer-service detection, graceful degradation if absent

Orchestration, tools, and governance

PrincipleSystem 1, test agentsSystem 2, project platform
Orchestrator without business logicOrchestrator use case that composes, with no business logic of its ownThin orchestration entry point, delegating to the registry
Tool registry by assemblyTools assembled and exposed by a dedicated builderSingle tool registry, centralized registration and resolution
Cross-cutting middleware chainTracing and approval envelope around every toolLogging and metrics middleware plus a data access guard
Tool filtering by accessContextual exposure of tools to the modelTool selection by access level, upstream
Human approval on the actionApproval request declared in the tool's contractData access guard before any mutation
Guardrails and budgetsSafety net around the tools, counters, rate limitRate limits, read-only access for generic queries

Execution, state, and observability

PrincipleSystem 1, test agentsSystem 2, project platform
Specialized capability as a sidecarExecution engine driven via the tool protocol, dedicated portExternal tool servers via a dedicated client
Exposed and guarded tool surfacePublished tool server, protected by a dedicated guardTool server integration, guard and limits
Observability through eventsLifecycle events traced, trace exportMetrics middleware, structured logs
External state and topologyPersistence behind its port, in-memory structures identifiedDedicated real-time state and migrations, orchestrated deployment

! Dynamic delegation is already under way

A notable point for the trajectory toward adaptive orchestration: in system 1, the orchestrator already has dynamic delegation building blocks. There is a sub-agent invoker, sub-agent templates, parallel and template-based delegation, a procedural memory that records and replays routines, and a behavior evolution subject to human validation, still held in to-review status. Tier 3 of maturity is therefore not a distant target: its foundation already exists.

! Memory, evaluation, budgets: where the evidence lives

The three reinforcements of this edition also rest on existing code. For memory (section 06), the third system, dedicated to long-term memory, already applies non-destructive forgetting: a periodic task updates the retrieval score without deleting memories, only derived links are pruned below a threshold, and importance as well as salience are produced as a by-product of asynchronous extraction, with no dedicated model call; validity over time, two timestamps per fact, is its natural extension. For evaluation (section 12), system 1 already has a memory evaluation bench, a trace export, a reliability feedback loop whose action remains deliberately manual, and a per-agent confidence score: the guarded runtime loop is their culmination. For distributed delegation (sections 04, 05, 07), system 1 already signs agent cards and caps the budget with an aggregated ceiling in autonomous mode; discovery by cards and block-based budget allocation are the distributed adapters of those same ports.

Talk to us

Tell us who you are, and we will get back to you.

Your details are recorded by Another Framework, publisher of runward, to get back to you. They are neither sold nor shared, and are deleted three years after our last exchange. To see or delete them: contact@runward.dev.