sdd:plan

Shared Instructions

• Transition Logging — append a transition entry on every .spec-context.json write
• Hook Execution — run user-configured hooks at supported pipeline points
• Layered Context — load Layer 1 living specs by domain (Step 1) and grade alignment (Step 2b)

Steps

1. Load Context

If $ARGUMENTS is provided, use specs/{$ARGUMENTS}/ as the target directory. Otherwise, find the most recently modified directory under specs/ that contains a spec.md.

Read in parallel:

• specs/{NNN}-{slug}/spec.md — feature name, requirements, scenarios
• specs/{NNN}-{slug}/.spec-context.json — current step/task (if exists)
• .sdd/principles.md — Layer 0 of the layered context model. Optional. If the file exists at the project root, load it; the bullet list of MUSTs will be checked against the Approach in Step 2 (Principles Check). If absent, skip the check entirely — principles.md is opt-in by presence.
• Layer 1 (Living specs). Per layered-context: if loadedDomains is already in .spec-context.json (set by /sdd:specify Step 3b), read each domain's .spec.md at its resolved path. Otherwise run python3 lib/scripts/resolve-spec-paths.py --changed <Files-to-Change> (or files referenced in the Summary) and persist the matched names to loadedDomains. Read in returned order (most-specific first). Empty list → skip Layer 1 silently.

If no spec found, stop: "Run /sdd:specify first."

Update specs/{NNN}-{slug}/.spec-context.json and append a transition entry per transition-logging:

{ "currentStep": "plan", "currentTask": null, "progress": "loading", "next": "tasks", "updated": "{TODAY}" }

Run pre:plan hooks per hook-execution with vars = { slug, spec-dir }.