ctx-cc 3.3.2 → 3.3.4

package/agents/ctx-verifier.md

@@ -399,9 +399,42 @@ Based on results:
 - Update PRD progress
 
 **If FAIL:**
-- Create fix tasks mapped to failing criteria
-- Set status = "debugging" or "executing"
-- Keep current story
+- Analyze failure type:
+  - **Build/Test/Runtime error** → Set status = "debugging"
+  - **Missing feature** → Set status = "executing" with fix tasks
+- Record `debug_issue` with error details
+- **DO NOT mark as complete until 100% pass**
+
+## 11. Delivery Guarantee Loop
+
+```
+┌─────────────────────────────────────────┐
+│         DELIVERY GUARANTEE LOOP         │
+├─────────────────────────────────────────┤
+│                                         │
+│  verify → ALL CRITERIA PASS?            │
+│     │                                   │
+│     ├─ YES → COMPLETE (100% working)    │
+│     │                                   │
+│     └─ NO → debug/fix → verify again    │
+│              ↑                     │    │
+│              └─────────────────────┘    │
+│                                         │
+│  Loop continues until:                  │
+│  1. ALL criteria pass, OR               │
+│  2. Max debug attempts (10) → escalate  │
+│                                         │
+│  NEVER mark complete with failures      │
+│                                         │
+└─────────────────────────────────────────┘
+```
+
+**Key principle:** Work is ONLY marked complete when:
+- ✓ ALL acceptance criteria pass
+- ✓ Three-level check passes (exists, substantive, wired)
+- ✓ No blocking anti-patterns
+- ✓ Browser verification passes (if UI)
+- ✓ Build + Tests + Lint all green
 
 </process>
 

package/commands/ctx.md

@@ -88,8 +88,12 @@ Route to: **Three-Level Verification**
    - Level 2: Substantive (real code, not stub?)
    - Level 3: Wired (imported and used?)
 3. Scan for anti-patterns (TODO, empty catch, placeholders)
-4. If all pass: complete phase, update STATE.md
-5. If fails: create fix tasks, set status = "executing"
+4. Browser verification (if UI)
+5. **DELIVERY GUARANTEE:**
+   - If ALL pass: complete phase → COMPLETE
+   - If ANY fail: set status = "debugging" → fix → verify again
+   - Loop until 100% working or max attempts
+   - NEVER mark complete with failures
 
 ### If status = "paused"
 Route to: **Resume**
@@ -149,9 +153,14 @@ executing → debugging (if verification fails)
 executing → verifying (if all tasks done)
 debugging → executing (if fix works)
 debugging → ESCALATE (if max attempts fail)
-verifying → executing (if anti-patterns found)
-verifying → COMPLETE (if all passes)
+verifying → debugging (if any check fails)    ← LOOP BACK
+verifying → COMPLETE (if ALL pass 100%)       ← ONLY EXIT
 paused → (previous state)
+
+DELIVERY GUARANTEE LOOP:
+  verifying ←→ debugging ←→ executing
+       ↓
+  COMPLETE (only when 100% verified)
 ```
 </state_transitions>
 

package/commands/help.md

@@ -13,7 +13,7 @@ Output ONLY the reference content below. Do NOT add project-specific analysis.
 # CTX 3.3 Command Reference
 
 **CTX** (Continuous Task eXecution) - Intelligent workflow orchestration for Claude Code.
-20 agents. Learning system. Predictive planning. Self-healing. Voice control.
+21 agents. Learning system. Predictive planning. Self-healing. Voice control. Full system QA.
 
 ## Quick Start
 
@@ -103,6 +103,17 @@ Output ONLY the reference content below. Do NOT add project-specific analysis.
 | `/ctx debug --status` | Show current session status |
 | `/ctx debug --abort` | Abort current session |
 
+### QA (Full System Testing)
+| Command | Purpose |
+|---------|---------|
+| `/ctx qa` | Full system QA - crawl all pages (WCAG 2.1 AA) |
+| `/ctx qa --pages` | List discovered pages first |
+| `/ctx qa --section "auth"` | QA specific section only |
+| `/ctx qa --a11y-only` | Accessibility audit only |
+| `/ctx qa --visual-only` | Visual regression (mobile/tablet/desktop) |
+| `/ctx qa --resume` | Resume interrupted QA |
+| `/ctx qa --report` | Show last QA report |
+
 ### Session
 | Command | Purpose |
 |---------|---------|