fix: plan skill requires explicit user review before execution

skills/plan/SKILL.md

@@ -177,10 +177,25 @@ Show the user a summary:
 > .loop/loop.sh              # Headless
 > ```
 
-### Step 8: Wait for Feedback
+### Step 8: Review Before Execution
 
-Ask the user:
-> **Ready to start?** I can begin implementing now, or you can adjust the plan first.
+Tell the user:
+> **Review the plan before running.** The stories and contracts are designed to be human-reviewed and adjusted before handing off to the autonomous loop.
+>
+> **Files to review:**
+> - `.loop/prd.json` — stories, acceptance criteria, priorities
+> - `.loop/contracts/` — sprint contracts with done conditions and scope
+>
+> **Common adjustments:**
+> - Split a story that's too large
+> - Reorder priorities
+> - Tighten or loosen acceptance criteria
+> - Add/remove stories
+> - Adjust scope in contracts
+>
+> Let me know what changes you'd like, or when you're happy with the plan, run:
+> ```
+> /agent-loop:run
+> ```
 
-- If the user confirms (yes, go, start, implement, etc.) → invoke `/agent-loop:run` to begin the loop immediately
-- If they want changes (split a story, reorder, add/remove, change criteria) → make the changes, re-present the summary, and ask again
+Wait for the user to review. If they request changes, make them and re-present the summary. Do NOT automatically start the loop — the user must explicitly invoke `/agent-loop:run` when they're ready.