feat: add dual-mode support to test-fix-gen workflow

Enable test generation for existing codebases without requiring prior
workflow session. Supports both session-based (WFS-xxx) and prompt-based
(text description or file path) inputs with automatic mode detection.

Key improvements:
- Prompt mode: Generate tests from feature description or requirements file
- Backward compatible: Existing session-based usage unchanged
- Flexible input: Accepts session ID, text description, or file path
- Use case: Testing existing APIs without workflow history

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>

.claude/commands/cli/codex-execute.md

@@ -284,10 +284,22 @@ codex exec "CONTINUE TO NEXT SUBTASK: ..." resume --last --skip-git-repo-check -
 1. Mark subtask as blocked in TodoWrite
 2. Report error details to user
 3. Pause execution for manual intervention
-4. User can choose to:
+4. Use AskUserQuestion for recovery decision:
-   - Retry current subtask
+
-   - Continue to next subtask
+```typescript
-   - Abort entire task
+AskUserQuestion({
+  questions: [{
+    question: "Codex execution failed for the subtask. How should the workflow proceed?",
+    header: "Recovery",
+    options: [
+      { label: "Retry Subtask", description: "Attempt to execute the same subtask again." },
+      { label: "Skip Subtask", description: "Continue to the next subtask in the plan." },
+      { label: "Abort Workflow", description: "Stop the entire execution." }
+    ],
+    multiSelect: false
+  }]
+})
+```
 
 **Git Verification Failure** (if --verify-git):
 1. Show unexpected changes

.claude/commands/cli/discuss-plan.md

@@ -121,9 +121,22 @@ Claude (orchestrating AI) synthesizes both outputs:
 ### Phase 3: User Review and Iteration
 
 1.  **Present Synthesis**: Show synthesized plan and key discussion points
-2.  **Continue or Conclude**: Prompt user:
+2.  **Continue or Conclude**: Use AskUserQuestion to prompt user:
-    - **(1)** Start another round of discussion
+
-    - **(2)** Conclude and finalize the plan
+```typescript
+AskUserQuestion({
+  questions: [{
+    question: "Round of discussion complete. What is the next step?",
+    header: "Next Round",
+    options: [
+      { label: "Start another round", description: "Continue the discussion to refine the plan further." },
+      { label: "Conclude and finalize", description: "End the discussion and save the final plan." }
+    ],
+    multiSelect: false
+  }]
+})
+```
+
 3.  **Loop or Finalize**:
     -   Continue → New round with Gemini analyzing latest synthesis
     -   Conclude → Save final synthesized document

.claude/commands/task/breakdown.md

@@ -68,7 +68,21 @@ Define subtasks manually (remaining capacity: 4 tasks):
    - "User authentication" and "OAuth integration" both handle auth
    - Consider combining into single task
 
-Proceed with breakdown? (y/n): y
+# Use AskUserQuestion for confirmation
+AskUserQuestion({
+  questions: [{
+    question: "File conflicts and/or similar functionality detected. How do you want to proceed?",
+    header: "Confirm",
+    options: [
+      { label: "Proceed with breakdown", description: "Accept the risks and create the subtasks as defined." },
+      { label: "Restart breakdown", description: "Discard current subtasks and start over." },
+      { label: "Cancel breakdown", description: "Abort the operation and leave the parent task as is." }
+    ],
+    multiSelect: false
+  }]
+})
+
+User selected: "Proceed with breakdown"
 
 ✅ Task IMPL-1 broken down:
 ▸ IMPL-1: Build authentication module (container)

.claude/commands/task/replan.md

@@ -174,7 +174,20 @@ Rollback to version 1.1:
 - Remove new subtasks if any
 - Update session stats
 
-Confirm rollback? (y/n): y
+# Use AskUserQuestion for confirmation
+AskUserQuestion({
+  questions: [{
+    question: "Are you sure you want to roll back this task to a previous version?",
+    header: "Confirm",
+    options: [
+      { label: "Yes, rollback", description: "Restore the task from the selected backup." },
+      { label: "No, cancel", description: "Keep the current version of the task." }
+    ],
+    multiSelect: false
+  }]
+})
+
+User selected: "Yes, rollback"
 
 ✅ Task rolled back to version 1.1
 ```
@@ -227,7 +240,20 @@ Proposed updates:
 + Add OAuth2 integration
 + Update authentication flow
 
-Apply changes? (y/n): y
+# Use AskUserQuestion for confirmation
+AskUserQuestion({
+  questions: [{
+    question: "Do you want to apply these changes to the task?",
+    header: "Apply",
+    options: [
+      { label: "Yes, apply", description: "Create new version with these changes." },
+      { label: "No, cancel", description: "Discard changes and keep current version." }
+    ],
+    multiSelect: false
+  }]
+})
+
+User selected: "Yes, apply"
 
 ✓ Version 1.2 created
 ✓ Context updated

.claude/commands/workflow/test-fix-gen.md

@@ -1,7 +1,7 @@
 ---
 name: test-fix-gen
-description: Create independent test-fix workflow session by analyzing completed implementation
+description: Create independent test-fix workflow session from existing implementation (session or prompt-based)
-argument-hint: "[--use-codex] [--cli-execute] source-session-id"
+argument-hint: "[--use-codex] [--cli-execute] (source-session-id | \"feature description\" | /path/to/file.md)"
 allowed-tools: SlashCommand(*), TodoWrite(*), Read(*), Bash(*)
 ---
 
@@ -9,13 +9,16 @@ allowed-tools: SlashCommand(*), TodoWrite(*), Read(*), Bash(*)
 
 ## Coordinator Role
 
-**This command is a pure orchestrator**: Creates an independent test-fix workflow session for validating a completed implementation. It reuses the standard planning toolchain with automatic cross-session context gathering.
+**This command is a pure orchestrator**: Creates an independent test-fix workflow session for existing code. Supports two input modes:
+- **Session Mode**: Analyze completed workflow session (cross-session context gathering)
+- **Prompt Mode**: Analyze existing codebase via description or file (prompt-based context gathering)
 
 **Core Principles**:
-- **Session Isolation**: Creates new `WFS-test-[source]` session to keep verification separate from implementation
+- **Dual Input Support**: Accepts either session ID (WFS-xxx) or feature description/file path
-- **Context-First**: Prioritizes gathering code changes and summaries from source session
+- **Session Isolation**: Creates new `WFS-test-[slug]` session to keep test workflow independent
-- **Format Reuse**: Creates standard `IMPL-*.json` task, using `meta.type: "test-fix"` for agent assignment
+- **Context-First**: Gathers implementation context via appropriate method (session or prompt-based)
-- **Parameter Simplification**: Tools auto-detect test session type via metadata, no manual cross-session parameters needed
+- **Format Reuse**: Creates standard `IMPL-*.json` tasks, using `meta.type: "test-fix"` for agent assignment
+- **Automatic Mode Detection**: Input pattern determines execution mode (no manual flags)
 - **Manual First**: Default to manual fixes, use `--use-codex` flag for automated Codex fix application
 
 **Execution Flow**:
@@ -339,3 +342,76 @@ See `/workflow:tools:test-task-generate` for detailed specifications.
 - `/workflow:test-cycle-execute` - Execute test-fix workflow with dynamic iteration
 - `/workflow:execute` - Execute standard workflow tasks
 - `/workflow:status` - Check progress
+
+## Dual-Mode Support (Enhanced)
+
+### Input Mode Detection
+
+**Automatic mode detection based on argument pattern**:
+
+```bash
+# Detection Logic
+if [[ "$input" == WFS-* ]]; then
+  MODE="session"  # Session Mode
+elif [ -f "$input" ]; then
+  MODE="prompt"   # Prompt Mode (file)
+else
+  MODE="prompt"   # Prompt Mode (text)
+fi
+```
+
+### Mode Comparison
+
+| Aspect | Session Mode | Prompt Mode |
+|--------|-------------|-------------|
+| **Input** | `WFS-xxx` pattern | Description or file path |
+| **Phase 1** | Create `WFS-test-[source]` with `source_session_id` | Create `WFS-test-[slug]` without `source_session_id` |
+| **Phase 2** | `/workflow:tools:test-context-gather` | `/workflow:tools:context-gather` |
+| **Phase 3-5** | Identical | Identical |
+| **Context Source** | Source session summaries | Direct codebase analysis |
+
+### Usage Examples
+
+#### Session Mode (Existing Behavior)
+```bash
+# Test validation for completed implementation session
+/workflow:test-fix-gen WFS-user-auth-v2
+```
+
+#### Prompt Mode - Text Description
+```bash
+# Generate tests from feature description
+/workflow:test-fix-gen "Test the user authentication API endpoints in src/auth/api.ts"
+```
+
+#### Prompt Mode - File Reference
+```bash
+# Generate tests from requirements file
+/workflow:test-fix-gen ./docs/api-requirements.md
+```
+
+#### With Codex Automation
+```bash
+# Session mode with automated fixes
+/workflow:test-fix-gen --use-codex WFS-user-auth
+
+# Prompt mode with automated fixes
+/workflow:test-fix-gen --use-codex "Test user registration flow"
+```
+
+### Implementation Notes
+
+**Core Rules Addition**:
+- Rule 0: **Detect Input Mode** - First analyze input argument to determine session vs prompt mode
+
+**Phase 1 Variation**:
+- Session: `"Test validation for [sourceSessionId]"`
+- Prompt: `"Test generation for: [prompt-description]"`
+
+**Phase 2 Variation**:
+- Session: `test-context-gather` (reads `source_session_id` from metadata)
+- Prompt: `context-gather --session [sessionId] "[task_description]"`
+
+### Backward Compatibility
+
+✅ **Fully backward compatible**: Existing session-based usage remains unchanged. All `WFS-*` arguments automatically use session mode.