component-model

name

component-model

description

Decides where new logic lives in the component model — agent, skill, command, KB, or template — and how to keep components in their layer, routing each decision to the right authoring resource and running the fat-to-thin refactor when a component has outgrown its layer. Loaded by the authoring flows (create-skill, create-agent, kb-build, create-kb) as their layer-decision gate. Use for "should this be an agent or a skill", "where does this logic go", adding a new component, "this agent is too big", or restructuring/refactoring an agent into skills. Do not use to author the chosen component itself — that is create-skill, create-agent, or /create-kb.

Component Model — layer decision and fat-to-thin refactor

The operational side of the repository's component model: place new logic in the right layer, and pull overgrown components back into theirs. The canonical definitions — the four layers, the placement table, the thin-executor pattern — live in .claude/kb/shared/component-model.md; this skill applies that model, it does not restate it.

When to use

• Deciding whether new functionality should be an agent, a skill, a command, a KB file, or a template.
• Planning a new component before authoring it — the authoring flows load this skill first as their layer-decision gate.
• A component has grown fat — an agent body full of process steps, a command re-implementing methodology inline — and needs restructuring.
• Reviewing a change that spans component types and looks misplaced.

Skip if

• The layer is already decided and you are authoring the component — go straight to create-skill, create-agent, or /create-kb.
• The change is content-only inside an existing component (fixing a step in a skill, updating a KB fact) — no layer question exists.

The decision procedure

1. Read the canonical model. Open — the four layers, the "where does new logic go" table, and the thin-executor pattern are defined there, and every classification below is grounded in that file. Read it before classifying, not after.