dx-plan-validate

You cross-check implement.md against explain.md to verify the plan is complete, correct, and ready for execution.

Use ultrathink for this skill — careful cross-referencing benefits from deep reasoning.

Output

You run in a forked context. Before emitting any chat output, determine whether you were invoked by the orchestrator (dx-agent-all) or standalone — see plugins/dx-core/shared/orchestration-check.md:

ORCHESTRATED=0
FLAG=".ai/run-context/orchestrating.flag"
if [ -f "$FLAG" ]; then
  AGE=$(( $(date +%s) - $(date -r "$FLAG" +%s) ))
  [ "$AGE" -lt 7200 ] && ORCHESTRATED=1
fi

• If $ORCHESTRATED == 1 (orchestrator path): write the full report to $SPEC_DIR/validation-report.md and emit ONLY the ## Return block to chat.
• If $ORCHESTRATED == 0 (standalone path): write the same report AND emit the human-friendly summary marked <!-- standalone-only --> below, followed by the ## Return block at the very end.

Per-phase progress lines during the run are allowed in both paths.

This skill writes its full report to $SPEC_DIR/validation-report.md (overall verdict, per-requirement mapping, reuse check, warnings). The orchestrator reads this file only on demand.

1. Locate the Spec Directory

SPEC_DIR=$(bash .ai/lib/dx-common.sh find-spec-dir $ARGUMENTS)

Read from $SPEC_DIR:

• explain.md (required)
• implement.md (required)
• research.md (if exists — for file existence verification)

If either explain.md or implement.md is missing, print which is missing and STOP.

2. Validation Checks

Run these checks sequentially using extended thinking:

Check 1: Requirement Coverage

For each numbered requirement in explain.md, find at least one step in implement.md that addresses it.

Report:

• ✅ Requirement N — covered by Step X
• ❌ Requirement N — NOT covered by any step

Check 2: No Scope Creep

For each step in implement.md, verify it maps to at least one requirement in explain.md.

Report:

• ✅ Step N — implements Requirement X
• ⚠️ Step N — not directly tied to a requirement (flag for review)