authority-delegation-garden-patch

Agentic Architecture

Agentic Architecture covers the design of systems where AI agents operate as augmentations within human workflows — configuration lifecycle, multi-agent coordination, persona design, and tool orchestration. The unifying question: how do you build a system scaffold that lets agents do more without eroding human authority?

This domain is distinct from AI/ML generally. It is not concerned with training, inference, or model capabilities. It is concerned with the structural decisions that determine how agents behave in a specific human context — how they receive instructions, maintain state, coordinate with other agents, and know when to escalate.

Scope

Covers: Agent configuration lifecycle (how rules, skills, and state are stored and protected), multi-agent coordination patterns (Groundskeeper-Gardener commission architecture, persona differentiation), tool orchestration, the boundary between task instruction and role specialization, and naming discipline for agentic pattern languages.

Does not cover: AI/ML model design, training, or inference. Foundation model capabilities and limitations. General software architecture patterns not specific to agentic contexts. The [[Deep Context Architecture]] domain covers the garden’s own knowledge architecture; Agentic Architecture covers the agent system architecture that operates within it.

Key Forms

Configuration and State

How agent configuration, state, and identity are stored, protected, and separated:

Multi-Agent Coordination

How agents with different roles work together:

Commission Architecture

How bounded work is delegated, verified, and returned:

Process Failure Modes

How structured agent processes fail to produce their intended output:

Agent Learning and Reflection

How agents accumulate and distill operational experience:

Architectural Models

Meta-level explanations of why agent systems converge:

Naming Discipline

Vocabulary

Boundaries

Open Inquiries

Candidate Forms

Patterns and concepts identified but not yet extracted:

Patterns

Models

Inquiries

Open Questions

Sources

Relations