warp-os 1.1.3 → 1.2.1

package/agents/warp-plan-onboarding.md

@@ -119,6 +119,8 @@ Shell commands use Unix syntax (Git Bash). Never use CMD (`dir`, `type`, `del`)
 
 ## AskUserQuestion
 
+**Flow: analysis first, then decision tool.** Present your full reasoning, trade-offs, and recommendations as conversational text — the user wants to read your thinking. Then cap it with AskUserQuestion to formalize the decision. **If you're composing a message with multiple options or "which approach?" language, you MUST end it with AskUserQuestion.** Never present options in prose without the tool.
+
 **Contract:**
 1. **Re-ground:** Project name, branch, current task. (1-2 sentences.)
 2. **Simplify:** Plain English a smart 16-year-old could follow.
@@ -140,9 +142,15 @@ Shell commands use Unix syntax (Git Bash). Never use CMD (`dir`, `type`, `del`)
 Format: `"Option name — X/10 🟢"` (or 🟡 or 🔴). In the label, not the description.
 Rate: 🟢 9-10 complete, 🟡 6-8 adequate, 🔴 1-5 shortcuts.
 
+**Pre-call checklist (verify before every AskUserQuestion invocation):**
+- ☐ Completeness scores in every option label
+- ☐ Recommended option listed first
+- ☐ One decision per question (split if multiple)
+- ☐ Analysis/reasoning already presented in message text above
+
 **Formatting:**
 - *Italics* for emphasis, not **bold** (bold for headers only).
-- After each answer: `✔ Decision {N} recorded [quicksave updated]`
+- After each answer: `✔ Decision {N} recorded`
 - Previews under 8 lines. Full mockups go in conversation text before the question.
 
 ---

package/agents/warp-plan-optimize.md

@@ -119,6 +119,8 @@ Shell commands use Unix syntax (Git Bash). Never use CMD (`dir`, `type`, `del`)
 
 ## AskUserQuestion
 
+**Flow: analysis first, then decision tool.** Present your full reasoning, trade-offs, and recommendations as conversational text — the user wants to read your thinking. Then cap it with AskUserQuestion to formalize the decision. **If you're composing a message with multiple options or "which approach?" language, you MUST end it with AskUserQuestion.** Never present options in prose without the tool.
+
 **Contract:**
 1. **Re-ground:** Project name, branch, current task. (1-2 sentences.)
 2. **Simplify:** Plain English a smart 16-year-old could follow.
@@ -140,9 +142,15 @@ Shell commands use Unix syntax (Git Bash). Never use CMD (`dir`, `type`, `del`)
 Format: `"Option name — X/10 🟢"` (or 🟡 or 🔴). In the label, not the description.
 Rate: 🟢 9-10 complete, 🟡 6-8 adequate, 🔴 1-5 shortcuts.
 
+**Pre-call checklist (verify before every AskUserQuestion invocation):**
+- ☐ Completeness scores in every option label
+- ☐ Recommended option listed first
+- ☐ One decision per question (split if multiple)
+- ☐ Analysis/reasoning already presented in message text above
+
 **Formatting:**
 - *Italics* for emphasis, not **bold** (bold for headers only).
-- After each answer: `✔ Decision {N} recorded [quicksave updated]`
+- After each answer: `✔ Decision {N} recorded`
 - Previews under 8 lines. Full mockups go in conversation text before the question.
 
 ---

package/agents/warp-plan-scope.md

@@ -119,6 +119,8 @@ Shell commands use Unix syntax (Git Bash). Never use CMD (`dir`, `type`, `del`)
 
 ## AskUserQuestion
 
+**Flow: analysis first, then decision tool.** Present your full reasoning, trade-offs, and recommendations as conversational text — the user wants to read your thinking. Then cap it with AskUserQuestion to formalize the decision. **If you're composing a message with multiple options or "which approach?" language, you MUST end it with AskUserQuestion.** Never present options in prose without the tool.
+
 **Contract:**
 1. **Re-ground:** Project name, branch, current task. (1-2 sentences.)
 2. **Simplify:** Plain English a smart 16-year-old could follow.
@@ -140,9 +142,15 @@ Shell commands use Unix syntax (Git Bash). Never use CMD (`dir`, `type`, `del`)
 Format: `"Option name — X/10 🟢"` (or 🟡 or 🔴). In the label, not the description.
 Rate: 🟢 9-10 complete, 🟡 6-8 adequate, 🔴 1-5 shortcuts.
 
+**Pre-call checklist (verify before every AskUserQuestion invocation):**
+- ☐ Completeness scores in every option label
+- ☐ Recommended option listed first
+- ☐ One decision per question (split if multiple)
+- ☐ Analysis/reasoning already presented in message text above
+
 **Formatting:**
 - *Italics* for emphasis, not **bold** (bold for headers only).
-- After each answer: `✔ Decision {N} recorded [quicksave updated]`
+- After each answer: `✔ Decision {N} recorded`
 - Previews under 8 lines. Full mockups go in conversation text before the question.
 
 ---

package/agents/warp-plan-security.md

@@ -119,6 +119,8 @@ Shell commands use Unix syntax (Git Bash). Never use CMD (`dir`, `type`, `del`)
 
 ## AskUserQuestion
 
+**Flow: analysis first, then decision tool.** Present your full reasoning, trade-offs, and recommendations as conversational text — the user wants to read your thinking. Then cap it with AskUserQuestion to formalize the decision. **If you're composing a message with multiple options or "which approach?" language, you MUST end it with AskUserQuestion.** Never present options in prose without the tool.
+
 **Contract:**
 1. **Re-ground:** Project name, branch, current task. (1-2 sentences.)
 2. **Simplify:** Plain English a smart 16-year-old could follow.
@@ -140,9 +142,15 @@ Shell commands use Unix syntax (Git Bash). Never use CMD (`dir`, `type`, `del`)
 Format: `"Option name — X/10 🟢"` (or 🟡 or 🔴). In the label, not the description.
 Rate: 🟢 9-10 complete, 🟡 6-8 adequate, 🔴 1-5 shortcuts.
 
+**Pre-call checklist (verify before every AskUserQuestion invocation):**
+- ☐ Completeness scores in every option label
+- ☐ Recommended option listed first
+- ☐ One decision per question (split if multiple)
+- ☐ Analysis/reasoning already presented in message text above
+
 **Formatting:**
 - *Italics* for emphasis, not **bold** (bold for headers only).
-- After each answer: `✔ Decision {N} recorded [quicksave updated]`
+- After each answer: `✔ Decision {N} recorded`
 - Previews under 8 lines. Full mockups go in conversation text before the question.
 
 ---

package/agents/warp-plan-testdesign.md

@@ -119,6 +119,8 @@ Shell commands use Unix syntax (Git Bash). Never use CMD (`dir`, `type`, `del`)
 
 ## AskUserQuestion
 
+**Flow: analysis first, then decision tool.** Present your full reasoning, trade-offs, and recommendations as conversational text — the user wants to read your thinking. Then cap it with AskUserQuestion to formalize the decision. **If you're composing a message with multiple options or "which approach?" language, you MUST end it with AskUserQuestion.** Never present options in prose without the tool.
+
 **Contract:**
 1. **Re-ground:** Project name, branch, current task. (1-2 sentences.)
 2. **Simplify:** Plain English a smart 16-year-old could follow.
@@ -140,9 +142,15 @@ Shell commands use Unix syntax (Git Bash). Never use CMD (`dir`, `type`, `del`)
 Format: `"Option name — X/10 🟢"` (or 🟡 or 🔴). In the label, not the description.
 Rate: 🟢 9-10 complete, 🟡 6-8 adequate, 🔴 1-5 shortcuts.
 
+**Pre-call checklist (verify before every AskUserQuestion invocation):**
+- ☐ Completeness scores in every option label
+- ☐ Recommended option listed first
+- ☐ One decision per question (split if multiple)
+- ☐ Analysis/reasoning already presented in message text above
+
 **Formatting:**
 - *Italics* for emphasis, not **bold** (bold for headers only).
-- After each answer: `✔ Decision {N} recorded [quicksave updated]`
+- After each answer: `✔ Decision {N} recorded`
 - Previews under 8 lines. Full mockups go in conversation text before the question.
 
 ---

package/agents/warp-qa-debug.md

@@ -119,6 +119,8 @@ Shell commands use Unix syntax (Git Bash). Never use CMD (`dir`, `type`, `del`)
 
 ## AskUserQuestion
 
+**Flow: analysis first, then decision tool.** Present your full reasoning, trade-offs, and recommendations as conversational text — the user wants to read your thinking. Then cap it with AskUserQuestion to formalize the decision. **If you're composing a message with multiple options or "which approach?" language, you MUST end it with AskUserQuestion.** Never present options in prose without the tool.
+
 **Contract:**
 1. **Re-ground:** Project name, branch, current task. (1-2 sentences.)
 2. **Simplify:** Plain English a smart 16-year-old could follow.
@@ -140,9 +142,15 @@ Shell commands use Unix syntax (Git Bash). Never use CMD (`dir`, `type`, `del`)
 Format: `"Option name — X/10 🟢"` (or 🟡 or 🔴). In the label, not the description.
 Rate: 🟢 9-10 complete, 🟡 6-8 adequate, 🔴 1-5 shortcuts.
 
+**Pre-call checklist (verify before every AskUserQuestion invocation):**
+- ☐ Completeness scores in every option label
+- ☐ Recommended option listed first
+- ☐ One decision per question (split if multiple)
+- ☐ Analysis/reasoning already presented in message text above
+
 **Formatting:**
 - *Italics* for emphasis, not **bold** (bold for headers only).
-- After each answer: `✔ Decision {N} recorded [quicksave updated]`
+- After each answer: `✔ Decision {N} recorded`
 - Previews under 8 lines. Full mockups go in conversation text before the question.
 
 ---

package/agents/warp-qa-test.md

@@ -119,6 +119,8 @@ Shell commands use Unix syntax (Git Bash). Never use CMD (`dir`, `type`, `del`)
 
 ## AskUserQuestion
 
+**Flow: analysis first, then decision tool.** Present your full reasoning, trade-offs, and recommendations as conversational text — the user wants to read your thinking. Then cap it with AskUserQuestion to formalize the decision. **If you're composing a message with multiple options or "which approach?" language, you MUST end it with AskUserQuestion.** Never present options in prose without the tool.
+
 **Contract:**
 1. **Re-ground:** Project name, branch, current task. (1-2 sentences.)
 2. **Simplify:** Plain English a smart 16-year-old could follow.
@@ -140,9 +142,15 @@ Shell commands use Unix syntax (Git Bash). Never use CMD (`dir`, `type`, `del`)
 Format: `"Option name — X/10 🟢"` (or 🟡 or 🔴). In the label, not the description.
 Rate: 🟢 9-10 complete, 🟡 6-8 adequate, 🔴 1-5 shortcuts.
 
+**Pre-call checklist (verify before every AskUserQuestion invocation):**
+- ☐ Completeness scores in every option label
+- ☐ Recommended option listed first
+- ☐ One decision per question (split if multiple)
+- ☐ Analysis/reasoning already presented in message text above
+
 **Formatting:**
 - *Italics* for emphasis, not **bold** (bold for headers only).
-- After each answer: `✔ Decision {N} recorded [quicksave updated]`
+- After each answer: `✔ Decision {N} recorded`
 - Previews under 8 lines. Full mockups go in conversation text before the question.
 
 ---

package/agents/warp-release-update.md

@@ -119,6 +119,8 @@ Shell commands use Unix syntax (Git Bash). Never use CMD (`dir`, `type`, `del`)
 
 ## AskUserQuestion
 
+**Flow: analysis first, then decision tool.** Present your full reasoning, trade-offs, and recommendations as conversational text — the user wants to read your thinking. Then cap it with AskUserQuestion to formalize the decision. **If you're composing a message with multiple options or "which approach?" language, you MUST end it with AskUserQuestion.** Never present options in prose without the tool.
+
 **Contract:**
 1. **Re-ground:** Project name, branch, current task. (1-2 sentences.)
 2. **Simplify:** Plain English a smart 16-year-old could follow.
@@ -140,9 +142,15 @@ Shell commands use Unix syntax (Git Bash). Never use CMD (`dir`, `type`, `del`)
 Format: `"Option name — X/10 🟢"` (or 🟡 or 🔴). In the label, not the description.
 Rate: 🟢 9-10 complete, 🟡 6-8 adequate, 🔴 1-5 shortcuts.
 
+**Pre-call checklist (verify before every AskUserQuestion invocation):**
+- ☐ Completeness scores in every option label
+- ☐ Recommended option listed first
+- ☐ One decision per question (split if multiple)
+- ☐ Analysis/reasoning already presented in message text above
+
 **Formatting:**
 - *Italics* for emphasis, not **bold** (bold for headers only).
-- After each answer: `✔ Decision {N} recorded [quicksave updated]`
+- After each answer: `✔ Decision {N} recorded`
 - Previews under 8 lines. Full mockups go in conversation text before the question.
 
 ---

package/agents/warp-setup.md

@@ -119,6 +119,8 @@ Shell commands use Unix syntax (Git Bash). Never use CMD (`dir`, `type`, `del`)
 
 ## AskUserQuestion
 
+**Flow: analysis first, then decision tool.** Present your full reasoning, trade-offs, and recommendations as conversational text — the user wants to read your thinking. Then cap it with AskUserQuestion to formalize the decision. **If you're composing a message with multiple options or "which approach?" language, you MUST end it with AskUserQuestion.** Never present options in prose without the tool.
+
 **Contract:**
 1. **Re-ground:** Project name, branch, current task. (1-2 sentences.)
 2. **Simplify:** Plain English a smart 16-year-old could follow.
@@ -140,9 +142,15 @@ Shell commands use Unix syntax (Git Bash). Never use CMD (`dir`, `type`, `del`)
 Format: `"Option name — X/10 🟢"` (or 🟡 or 🔴). In the label, not the description.
 Rate: 🟢 9-10 complete, 🟡 6-8 adequate, 🔴 1-5 shortcuts.
 
+**Pre-call checklist (verify before every AskUserQuestion invocation):**
+- ☐ Completeness scores in every option label
+- ☐ Recommended option listed first
+- ☐ One decision per question (split if multiple)
+- ☐ Analysis/reasoning already presented in message text above
+
 **Formatting:**
 - *Italics* for emphasis, not **bold** (bold for headers only).
-- After each answer: `✔ Decision {N} recorded [quicksave updated]`
+- After each answer: `✔ Decision {N} recorded`
 - Previews under 8 lines. Full mockups go in conversation text before the question.
 
 ---

package/agents/warp-upgrade.md

@@ -119,6 +119,8 @@ Shell commands use Unix syntax (Git Bash). Never use CMD (`dir`, `type`, `del`)
 
 ## AskUserQuestion
 
+**Flow: analysis first, then decision tool.** Present your full reasoning, trade-offs, and recommendations as conversational text — the user wants to read your thinking. Then cap it with AskUserQuestion to formalize the decision. **If you're composing a message with multiple options or "which approach?" language, you MUST end it with AskUserQuestion.** Never present options in prose without the tool.
+
 **Contract:**
 1. **Re-ground:** Project name, branch, current task. (1-2 sentences.)
 2. **Simplify:** Plain English a smart 16-year-old could follow.
@@ -140,9 +142,15 @@ Shell commands use Unix syntax (Git Bash). Never use CMD (`dir`, `type`, `del`)
 Format: `"Option name — X/10 🟢"` (or 🟡 or 🔴). In the label, not the description.
 Rate: 🟢 9-10 complete, 🟡 6-8 adequate, 🔴 1-5 shortcuts.
 
+**Pre-call checklist (verify before every AskUserQuestion invocation):**
+- ☐ Completeness scores in every option label
+- ☐ Recommended option listed first
+- ☐ One decision per question (split if multiple)
+- ☐ Analysis/reasoning already presented in message text above
+
 **Formatting:**
 - *Italics* for emphasis, not **bold** (bold for headers only).
-- After each answer: `✔ Decision {N} recorded [quicksave updated]`
+- After each answer: `✔ Decision {N} recorded`
 - Previews under 8 lines. Full mockups go in conversation text before the question.
 
 ---
@@ -311,11 +319,20 @@ WARP │ UPGRADE
   Hooks:     patched ✓
   claude-mem: installed ✓
   Stale:     {N items cleaned / none}
-  Restart:   required
+  Restart:   {context-aware — see below}
 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
 ```
 
-**Always remind:** "Restart Claude Code to pick up the new skills and hooks."
+### Restart guidance (context-aware)
+
+Skills are discovered dynamically — new skills are available immediately after install, no restart needed. Hooks fire on session lifecycle events (SessionStart, Stop, PreToolUse) — if hook *scripts* changed in this upgrade, a restart picks up the new behavior.
+
+Determine restart need:
+- **Hooks changed** (new hooks added, existing hooks modified) → `Restart: recommended (hooks updated)`
+- **Only skills/agents changed** → `Restart: not needed — new skills available now`
+- **Already current** (same version) → `Restart: not needed`
+
+When hooks changed, say: "Restart Claude Code to pick up updated hooks." When they didn't, say: "No restart needed — new skills are available now."
 
 ---
 
@@ -327,7 +344,7 @@ WARP │ UPGRADE
 - Check for and offer to clean stale artifacts
 - Verify claude-mem is installed
 - Show changelog summary
-- Remind user to restart Claude Code
+- Give context-aware restart guidance (hooks changed → recommend restart; skills only → no restart needed)
 
 **MUST NOT:**
 - Use git clone/pull to upgrade (npm handles distribution now)

package/bin/hooks/CLAUDE.md

@@ -0,0 +1,24 @@
+# Hooks
+
+Warp hook scripts for Claude Code's hook system. Each script handles a specific lifecycle event.
+
+## Hook scripts
+
+- **identity-foundation.sh** — SessionStart. Injects the Warp engineering foundation identity. Runs on all session sources.
+- **identity-briefing.sh** — SessionStart. Injects pipeline state, branch, P1 priorities via `additionalContext` JSON. Dual-path artifact lookup: checks `.warp/reports/planning/` then `.warp/pipeline/`.
+- **consistency-check.sh** — Stop. Validates CLAUDE.md and TODOS.md consistency on session end. Warns only (stdout), cannot block.
+- **validate-askuser.sh** — PreToolUse on AskUserQuestion. L1 deterministic gate that blocks calls if option labels lack completeness scores (`X/10` + emoji). Exit 2 = block.
+
+## Shared utilities
+
+- **_warp_json.sh** — JSON field extraction without jq. Source this in every hook. Provides `_warp_read_input`, `_warp_field`, `_warp_field_raw`, `_warp_escape_json`.
+- **_warp_html.sh** — Markdown-to-HTML conversion with Warp styling. Used by orchestrator for artifact preview.
+
+## Conventions
+
+- All scripts use `set -euo pipefail`. Every `grep` in a pipeline must have `|| true` to survive no-match (exit 1 + pipefail = crash).
+- Source `_warp_json.sh` and call `_warp_read_input` before extracting fields.
+- Exit codes: `0` = allow/success, `2` = block with message (PreToolUse only). Never exit 1 (treated as error, not block).
+- SessionStart hooks output JSON: `{"hookSpecificOutput":{"hookEventName":"SessionStart","additionalContext":"..."}}`. Escape newlines via `_warp_escape_json`.
+- Stop hooks can only warn via stdout. They cannot block session end.
+- Keep hooks fast (< 100ms). No network calls, no heavy computation.

package/bin/hooks/_warp_json.sh

@@ -22,6 +22,7 @@ _warp_read_input() {
 # Extract a JSON string field: "key": "value" → value
 # Handles optional whitespace around the colon.
 # Returns empty string if field not found.
+# Safe under set -eo pipefail (grep no-match → empty, not crash).
 _warp_field() {
   local key="$1"
   if [ -z "$_WARP_HOOK_INPUT" ]; then
@@ -31,11 +32,12 @@ _warp_field() {
   echo "$_WARP_HOOK_INPUT" | \
     grep -o "\"$key\"[[:space:]]*:[[:space:]]*\"[^\"]*\"" | \
     head -1 | \
-    sed "s/\"$key\"[[:space:]]*:[[:space:]]*\"//;s/\"$//"
+    sed "s/\"$key\"[[:space:]]*:[[:space:]]*\"//;s/\"$//" || true
 }
 
 # Extract a non-string JSON field: "key": true → true, "key": 42 → 42
 # For booleans, numbers, null. Returns empty string if not found.
+# Safe under set -eo pipefail (grep no-match → empty, not crash).
 _warp_field_raw() {
   local key="$1"
   if [ -z "$_WARP_HOOK_INPUT" ]; then
@@ -46,7 +48,7 @@ _warp_field_raw() {
     grep -o "\"$key\"[[:space:]]*:[[:space:]]*[^,}\"][^,}]*" | \
     head -1 | \
     sed "s/\"$key\"[[:space:]]*:[[:space:]]*//" | \
-    tr -d '[:space:]'
+    tr -d '[:space:]' || true
 }
 
 # Escape a string for safe embedding inside a JSON string value.

package/bin/hooks/identity-briefing.sh

@@ -1,7 +1,8 @@
 #!/usr/bin/env bash
 # identity-briefing.sh — Inject pipeline state + priorities into Claude's context
-# Fires on SessionStart (all sources). Bounded to ~100-200 tokens.
+# Fires on SessionStart (all sources including compact). Bounded to ~100-200 tokens.
 # Outputs partial briefing if some files are missing (graceful degradation).
+# Must survive set -eo pipefail — every pipeline has || true or || echo fallback.
 set -euo pipefail
 
 SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
@@ -19,28 +20,28 @@ RULE="━━━━━━━━━━━━━━━━━━━━━━━━
 
 # Branch and git state
 BRANCH=$(git branch --show-current 2>/dev/null || echo "unknown")
-UNCOMMITTED=$(git status --porcelain 2>/dev/null | wc -l | tr -d '[:space:]')
+UNCOMMITTED=$(git status --porcelain 2>/dev/null | wc -l | tr -d '[:space:]' || echo "0")
 if [ "$UNCOMMITTED" = "0" ]; then
   GIT_STATE="clean"
 else
   GIT_STATE="${UNCOMMITTED} files"
 fi
 
-# Pipeline state
+# Pipeline state — check new path (.warp/reports/planning/) then fall back to old (.warp/pipeline/)
 PIPELINE=""
 for artifact in brainstorm onboarding scope architecture design testspec security README; do
-  f=".warp/pipeline/${artifact}.md"
-  if [ -f "$f" ]; then
+  if [ -f ".warp/reports/planning/${artifact}.md" ] || [ -f ".warp/pipeline/${artifact}.md" ]; then
     PIPELINE="${PIPELINE} ✓${artifact}"
   else
     PIPELINE="${PIPELINE} ○${artifact}"
   fi
 done
 
-# Mode detection
-if [ -f ".warp/pipeline/README.md" ]; then
+# Mode detection — check new path then old
+if [ -f ".warp/reports/roadmap/README.md" ] || [ -f ".warp/pipeline/README.md" ]; then
   WARP_MODE="pipeline"
-elif [ -d ".warp/pipeline" ] && ls .warp/pipeline/*.md >/dev/null 2>&1; then
+elif ([ -d ".warp/reports/planning" ] && ls .warp/reports/planning/*.md >/dev/null 2>&1) || \
+     ([ -d ".warp/pipeline" ] && ls .warp/pipeline/*.md >/dev/null 2>&1); then
   WARP_MODE="pipeline-partial"
 else
   WARP_MODE="no-pipeline"
@@ -53,11 +54,17 @@ if [ -n "$PIPELINE" ]; then
   BRIEFING="${BRIEFING}"$'\n'"  Pipeline:${PIPELINE}"
 fi
 
-# Phase/cycle progress from roadmap
-if [ -f ".warp/pipeline/README.md" ]; then
-  CHECKED=$(grep -c '^\- \[x\]' .warp/pipeline/README.md 2>/dev/null || echo "0")
-  TOTAL=$(grep -c '^\- \[' .warp/pipeline/README.md 2>/dev/null || echo "0")
-  NEXT=$(grep -m1 '^\- \[ \]' .warp/pipeline/README.md 2>/dev/null | sed 's/^- \[ \] //' | head -c 60 || true)
+# Phase/cycle progress from roadmap — check new path then old
+ROADMAP=""
+if [ -f ".warp/reports/roadmap/README.md" ]; then
+  ROADMAP=".warp/reports/roadmap/README.md"
+elif [ -f ".warp/pipeline/README.md" ]; then
+  ROADMAP=".warp/pipeline/README.md"
+fi
+if [ -n "$ROADMAP" ]; then
+  CHECKED=$(grep -c '^\- \[x\]' "$ROADMAP" 2>/dev/null || echo "0")
+  TOTAL=$(grep -c '^\- \[' "$ROADMAP" 2>/dev/null || echo "0")
+  NEXT=$(grep -m1 '^\- \[ \]' "$ROADMAP" 2>/dev/null | sed 's/^- \[ \] //' | head -c 60 || true)
   if [ "$TOTAL" != "0" ]; then
     BRIEFING="${BRIEFING}"$'\n'"Cycles: ${CHECKED}/${TOTAL} complete"
     if [ -n "$NEXT" ]; then

package/bin/hooks/validate-askuser.sh

@@ -0,0 +1,41 @@
+#!/usr/bin/env bash
+# validate-askuser.sh — PreToolUse gate for AskUserQuestion
+# Blocks the call if any option label is missing completeness scores.
+# Pattern required: "X/10" followed by 🟢, 🟡, or 🔴 somewhere in the label.
+#
+# L1 enforcement — deterministic regex, zero AI judgment.
+# Exit 0 = allow, exit 2 = block with message.
+set -euo pipefail
+
+SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
+source "$SCRIPT_DIR/_warp_json.sh"
+_warp_read_input
+
+# Extract all label values from the tool input.
+# Labels appear as "label": "Option name — 8/10 🟡" in the JSON.
+LABELS=$(echo "$_WARP_HOOK_INPUT" | grep -o '"label"[[:space:]]*:[[:space:]]*"[^"]*"' | sed 's/"label"[[:space:]]*:[[:space:]]*"//;s/"$//' || true)
+
+if [ -z "$LABELS" ]; then
+  # No labels found — might be malformed, let it through
+  exit 0
+fi
+
+# Check each label for the completeness pattern: digit(s)/10 + emoji
+MISSING=""
+while IFS= read -r label; do
+  # Match: one or more digits, /10, then anywhere in the label one of the three emojis
+  if ! echo "$label" | grep -qE '[0-9]+/10'; then
+    MISSING="${MISSING}\n  - \"${label}\""
+  fi
+done <<< "$LABELS"
+
+if [ -n "$MISSING" ]; then
+  echo "BLOCKED: Completeness scores missing from option labels." >&2
+  echo "Every AskUserQuestion option label MUST include 'X/10' + 🟢🟡🔴." >&2
+  echo "Labels missing scores:${MISSING}" >&2
+  echo "" >&2
+  echo "Format: \"Option name — X/10 🟢\" (🟢 9-10, 🟡 6-8, 🔴 1-5)" >&2
+  exit 2
+fi
+
+exit 0