feat: Optimize Codex prompt commands parameter flexibility

- Enhanced 14 commands with flexible parameter support
- Standardized argument formats across all commands
- Added English parameter descriptions for clarity
- Maintained backward compatibility

Commands optimized:
- analyze-with-file: Added --depth, --max-iterations
- brainstorm-with-file: Added --perspectives, --max-ideas, --focus
- debug-with-file: Added --scope, --focus, --depth
- issue-execute: Unified format, added --skip-tests, --skip-build, --dry-run
- lite-plan-a/b/c: Added depth and execution control flags
- execute: Added --parallel, --filter, --skip-tests
- brainstorm-to-cycle: Unified to --session format, added --launch
- lite-fix: Added --hotfix, --severity, --scope
- clean: Added --focus, --target, --confirm
- lite-execute: Unified --plan format, added execution control
- compact: Added --description, --tags, --force
- issue-new: Complete flexible parameter support

Unchanged (already optimal):
- issue-plan, issue-discover, issue-queue, issue-discover-by-prompt

.codex/prompts/analyze-with-file.md

@@ -1,6 +1,6 @@
 ---
-description: Interactive collaborative analysis with documented discussions, CLI-assisted exploration, and evolving understanding
+description: Interactive collaborative analysis with documented discussions, CLI-assisted exploration, and evolving understanding. Supports depth control and iteration limits.
-argument-hint: TOPIC="<topic or question to analyze>"
+argument-hint: "TOPIC=\"<topic or question>\" [--depth=standard|deep|full] [--max-iterations=<n>] [--verbose]"
 ---
 
 # Codex Analyze-With-File Prompt
@@ -22,6 +22,9 @@ Interactive collaborative analysis workflow with **documented discussion process
 
 **$TOPIC**
 
+- `--depth`: Analysis depth (standard|deep|full)
+- `--max-iterations`: Max discussion rounds
+
 ## Execution Process
 
 ```

.codex/prompts/brainstorm-to-cycle.md

@@ -1,6 +1,6 @@
 ---
-description: Convert brainstorm session output to parallel-dev-cycle input with idea selection and context enrichment
+description: Convert brainstorm session output to parallel-dev-cycle input with idea selection and context enrichment. Unified parameter format.
-argument-hint: SESSION="<brainstorm-session-id>" [--idea=<index>] [--auto]
+argument-hint: "--session=<id> [--idea=<index>] [--auto] [--launch]"
 ---
 
 # Brainstorm to Cycle Adapter
@@ -15,9 +15,10 @@ Bridge workflow that converts **brainstorm-with-file** output to **parallel-dev-
 
 | Argument | Required | Description |
 |----------|----------|-------------|
-| SESSION | Yes | Brainstorm session ID (e.g., `BS-rate-limiting-2025-01-28`) |
+| --session | Yes | Brainstorm session ID (e.g., `BS-rate-limiting-2025-01-28`) |
 | --idea | No | Pre-select idea by index (0-based, from top_ideas) |
 | --auto | No | Auto-select top-scored idea without confirmation |
+| --launch | No | Auto-launch parallel-dev-cycle without preview |
 
 ## Output
 

.codex/prompts/brainstorm-with-file.md

@@ -1,6 +1,6 @@
 ---
-description: Interactive brainstorming with multi-perspective analysis, idea expansion, and documented thought evolution
+description: Interactive brainstorming with multi-perspective analysis, idea expansion, and documented thought evolution. Supports perspective selection and idea limits.
-argument-hint: TOPIC="<idea or topic to brainstorm>"
+argument-hint: "TOPIC=\"<idea or topic>\" [--perspectives=role1,role2,...] [--max-ideas=<n>] [--focus=<area>] [--verbose]"
 ---
 
 # Codex Brainstorm-With-File Prompt
@@ -22,6 +22,10 @@ Interactive brainstorming workflow with **documented thought evolution**. Expand
 
 **$TOPIC**
 
+- `--perspectives`: Analysis perspectives (role1,role2,...)
+- `--max-ideas`: Max number of ideas
+- `--focus`: Focus area
+
 ## Execution Process
 
 ```

.codex/prompts/clean.md

@@ -1,6 +1,6 @@
 ---
-description: Intelligent code cleanup with mainline detection, stale artifact discovery, and safe execution
+description: Intelligent code cleanup with mainline detection, stale artifact discovery, and safe execution. Supports targeted cleanup and confirmation.
-argument-hint: [--dry-run] [FOCUS="<area>"]
+argument-hint: "[--dry-run] [--focus=<area>] [--target=sessions|documents|dead-code] [--confirm]"
 ---
 
 # Workflow Clean Command
@@ -16,6 +16,11 @@ Evidence-based intelligent cleanup command. Systematically identifies stale arti
 **Focus area**: $FOCUS (or entire project if not specified)
 **Mode**: $ARGUMENTS
 
+- `--dry-run`: Preview cleanup without executing
+- `--focus`: Focus area (module or path)
+- `--target`: Cleanup target (sessions|documents|dead-code)
+- `--confirm`: Skip confirmation, execute directly
+
 ## Execution Process
 
 ```

.codex/prompts/compact.md

@@ -1,6 +1,6 @@
 ---
-description: Compact current session memory into structured text for session recovery
+description: Compact current session memory into structured text for session recovery. Supports custom descriptions and tagging.
-argument-hint: "[optional: session description]"
+argument-hint: "[--description=\"...\"] [--tags=<tag1,tag2>] [--force]"
 ---
 
 # Memory Compact Command (/memory:compact)
@@ -17,9 +17,11 @@ The `memory:compact` command **compresses current session working memory** into
 
 ## 2. Parameters
 
-- `"session description"` (Optional): Session description to supplement objective
+- `--description`: Custom session description (optional)
   - Example: "completed core-memory module"
   - Example: "debugging JWT refresh - suspected memory leak"
+- `--tags`: Comma-separated tags for categorization (optional)
+- `--force`: Skip confirmation, save directly
 
 ## 3. Structured Output Format
 

.codex/prompts/debug-with-file.md

@@ -1,6 +1,6 @@
 ---
-description: Interactive hypothesis-driven debugging with documented exploration, understanding evolution, and analysis-assisted correction
+description: Interactive hypothesis-driven debugging with documented exploration, understanding evolution, and analysis-assisted correction. Supports scope and focus control.
-argument-hint: BUG="<bug description or error message>"
+argument-hint: "BUG=\"<bug description or error message>\" [--scope=<path>] [--focus=<component>] [--depth=standard|deep] [--verbose]"
 ---
 
 # Codex Debug-With-File Prompt
@@ -21,6 +21,10 @@ Enhanced evidence-based debugging with **documented exploration process**. Recor
 
 **$BUG**
 
+- `--scope`: Debug scope limit (file path)
+- `--focus`: Focus component
+- `--depth`: Debug depth (standard|deep)
+
 ## Execution Process
 
 ```

.codex/prompts/execute.md

@@ -1,6 +1,6 @@
 ---
-description: Execute workflow tasks sequentially from session folder
+description: Execute workflow tasks sequentially from session folder. Supports parallel execution and task filtering.
-argument-hint: SESSION=<path-to-session-folder>
+argument-hint: "SESSION=<path-to-session-folder> [--parallel] [--filter=<pattern>] [--skip-tests]"
 ---
 
 # Workflow Execute (Codex Version)
@@ -13,6 +13,10 @@ argument-hint: SESSION=<path-to-session-folder>
 
 Session folder path via `$SESSION` (e.g., `.workflow/active/WFS-auth-system`)
 
+- `--parallel`: Execute tasks in parallel (default: sequential)
+- `--filter`: Filter tasks by pattern (e.g., `IMPL-1.*`)
+- `--skip-tests`: Skip test execution
+
 ## Task Tracking (JSON Source of Truth + Codex TODO Tool)
 
 - **Source of truth**: Task state MUST be read from and written to `$SESSION/.task/IMPL-*.json`.

.codex/prompts/issue-execute.md

@@ -1,6 +1,6 @@
 ---
-description: Execute all solutions from issue queue with git commit after each solution
+description: Execute all solutions from issue queue with git commit after each solution. Supports batch processing and execution control.
-argument-hint: "--queue <queue-id> [--worktree [<existing-path>]]"
+argument-hint: "--queue=<id> [--worktree=<path|new>] [--skip-tests] [--skip-build] [--dry-run] [--verbose]"
 ---
 
 # Issue Execute (Codex Version)
@@ -19,6 +19,14 @@ Before starting execution, load project context:
 
 This ensures execution follows project conventions and patterns.
 
+## Parameters
+
+- `--queue=<id>`: Queue ID to execute (REQUIRED)
+- `--worktree=<path|new>`: Worktree path or 'new' for creating new worktree
+- `--skip-tests`: Skip test execution during solution implementation
+- `--skip-build`: Skip build step
+- `--dry-run`: Preview execution without making changes
+
 ## Queue ID Requirement (MANDATORY)
 
 **`--queue <queue-id>` parameter is REQUIRED**

.codex/prompts/issue-new.md

@@ -1,19 +1,38 @@
 ---
-description: Create structured issue from GitHub URL or text description
+description: Create structured issue from GitHub URL or text description. Auto mode with --yes flag.
-argument-hint: "<github-url | text-description> [--priority 1-5]"
+argument-hint: "[--yes|-y] <GITHUB_URL | TEXT_DESCRIPTION> [--priority PRIORITY] [--labels LABELS]"
 ---
 
-# Issue New 
+# Issue New Command
 
-## Goal
+## Core Principles
 
-Create a new issue from a GitHub URL or text description. Detect input clarity and ask clarifying questions only when necessary. Register the issue for planning.
+**Requirement Clarity Detection** → Ask only when needed
-
+**Flexible Parameter Input** → Support multiple formats and flags
-**Core Principle**: Requirement Clarity Detection → Ask only when needed
+**Auto Mode Support** → `--yes`/`-y` skips confirmation questions
 
 ```
-Clear Input (GitHub URL, structured text)  → Direct creation
+Clear Input (GitHub URL, structured text)     → Direct creation (no questions)
-Unclear Input (vague description)          → Minimal clarifying questions
+Unclear Input (vague description)             → Clarifying questions (unless --yes)
+Auto Mode (--yes or -y flag)                  → Skip all questions, use inference
+```
+
+## Parameter Formats
+
+```bash
+# GitHub URL (auto-detected)
+/prompts:issue-new https://github.com/owner/repo/issues/123
+/prompts:issue-new GH-123
+
+# Text description with priority
+/prompts:issue-new "Login fails with special chars" --priority 1
+
+# Auto mode - skip all questions
+/prompts:issue-new --yes "something broken"
+/prompts:issue-new -y https://github.com/owner/repo/issues/456
+
+# With labels
+/prompts:issue-new "Database migration needed" --priority 2 --labels "enhancement,database"
 ```
 
 ## Issue Structure
@@ -78,25 +97,46 @@ echo '{"title":"...", "context":"...", "priority":3}' | ccw issue create
 
 ## Workflow
 
-### Step 1: Analyze Input Clarity
+### Phase 0: Parse Arguments & Flags
 
-Parse and detect input type:
+Extract parameters from user input:
+
+```bash
+# Input examples (Codex placeholders)
+INPUT="$1"          # GitHub URL or text description
+AUTO_MODE="$2"      # Check for --yes or -y flag
+
+# Parse flags (comma-separated in single argument)
+PRIORITY=$(echo "$INPUT" | grep -oP '(?<=--priority\s)\d+' || echo "3")
+LABELS=$(echo "$INPUT" | grep -oP '(?<=--labels\s)\K[^-]*' | xargs)
+AUTO_YES=$(echo "$INPUT" | grep -qE '--yes|-y' && echo "true" || echo "false")
+
+# Extract main input (URL or text) - remove all flags
+MAIN_INPUT=$(echo "$INPUT" | sed 's/\s*--priority\s*\d*//; s/\s*--labels\s*[^-]*//; s/\s*--yes\s*//; s/\s*-y\s*//' | xargs)
+```
+
+### Phase 1: Analyze Input & Clarity Detection
 
 ```javascript
-// Detection patterns
+const mainInput = userInput.trim();
-const isGitHubUrl = input.match(/github\.com\/[\w-]+\/[\w-]+\/issues\/\d+/);
+
-const isGitHubShort = input.match(/^#(\d+)$/);
+// Detect input type and clarity
-const hasStructure = input.match(/(expected|actual|affects|steps):/i);
+const isGitHubUrl = mainInput.match(/github\.com\/[\w-]+\/[\w-]+\/issues\/\d+/);
+const isGitHubShort = mainInput.match(/^GH-?\d+$/);
+const hasStructure = mainInput.match(/(expected|actual|affects|steps):/i);
 
 // Clarity score: 0-3
 let clarityScore = 0;
 if (isGitHubUrl || isGitHubShort) clarityScore = 3;  // GitHub = fully clear
 else if (hasStructure) clarityScore = 2;             // Structured text = clear
-else if (input.length > 50) clarityScore = 1;        // Long text = somewhat clear
+else if (mainInput.length > 50) clarityScore = 1;    // Long text = somewhat clear
 else clarityScore = 0;                               // Vague
+
+// Auto mode override: if --yes/-y flag, skip all questions
+const skipQuestions = process.env.AUTO_YES === 'true';
 ```
 
-### Step 2: Extract Issue Data
+### Phase 2: Extract Issue Data & Priority
 
 **For GitHub URL/Short:**
 
@@ -104,13 +144,14 @@ else clarityScore = 0;                               // Vague
 # Fetch issue details via gh CLI
 gh issue view <issue-ref> --json number,title,body,labels,url
 
-# Parse response
+# Parse response with priority override
 {
   "id": "GH-123",
   "title": "...",
+  "priority": $PRIORITY || 3,  # Use --priority flag if provided
   "source": "github",
   "source_url": "https://github.com/...",
-  "labels": ["bug", "priority:high"],
+  "labels": $LABELS || [...existing labels],
   "context": "..."
 }
 ```
@@ -126,10 +167,12 @@ const expected = text.match(/expected:?\s*([^.]+)/i);
 const actual = text.match(/actual:?\s*([^.]+)/i);
 const affects = text.match(/affects?:?\s*([^.]+)/i);
 
-// Build issue data
+// Build issue data with flags
 {
   "id": id,
   "title": text.split(/[.\n]/)[0].substring(0, 60),
+  "priority": $PRIORITY || 3,            # From --priority flag
+  "labels": $LABELS?.split(',') || [],   # From --labels flag
   "source": "text",
   "context": text.substring(0, 500),
   "expected_behavior": expected?.[1]?.trim(),
@@ -137,7 +180,7 @@ const affects = text.match(/affects?:?\s*([^.]+)/i);
 }
 ```
 
-### Step 3: Context Hint (Conditional)
+### Phase 3: Context Hint (Conditional)
 
 For medium clarity (score 1-2) without affected components:
 
@@ -150,11 +193,13 @@ Add discovered files to `affected_components` (max 3 files).
 
 **Note**: Skip this for GitHub issues (already have context) and vague inputs (needs clarification first).
 
-### Step 4: Clarification (Only if Unclear)
+### Phase 4: Conditional Clarification (Skip if Auto Mode)
 
-**Only for clarity score < 2:**
+**Only ask if**: clarity < 2 AND NOT in auto mode (skipQuestions = false)
 
-Present a prompt asking for more details:
+If auto mode (`--yes`/`-y`), proceed directly to creation with inferred details.
+
+Otherwise, present minimal clarification:
 
 ```
 Input unclear. Please describe:
@@ -165,17 +210,23 @@ Input unclear. Please describe:
 
 Wait for user response, then update issue data.
 
-### Step 5: GitHub Publishing Decision
+### Phase 5: GitHub Publishing Decision (Skip if Already GitHub)
 
 For non-GitHub sources, determine if user wants to publish to GitHub:
 
+```
+
+For non-GitHub sources AND NOT auto mode, ask:
+
 ```
 Would you like to publish this issue to GitHub?
 1. Yes, publish to GitHub (create issue and link it)
 2. No, keep local only (store without GitHub sync)
 ```
 
-### Step 6: Create Issue
+In auto mode: Default to NO (keep local only, unless explicitly requested with --publish flag).
+
+### Phase 6: Create Issue
 
 **Create via CLI:**
 
@@ -198,7 +249,7 @@ GH_NUMBER=$(echo $GH_URL | grep -oE '/issues/([0-9]+)$' | grep -oE '[0-9]+')
 ccw issue update ${ISSUE_ID} --github-url "${GH_URL}" --github-number ${GH_NUMBER}
 ```
 
-### Step 7: Output Result
+### Phase 7: Output Result
 
 ```markdown
 ## Issue Created
@@ -244,20 +295,96 @@ Before completing, verify:
 
 ## Start Execution
 
-Parse input and detect clarity:
+### Parameter Parsing (Phase 0)
 
 ```bash
-# Get input from arguments
+# Codex passes full input as $1
-INPUT="${1}"
+INPUT="$1"
 
-# Detect if GitHub URL
+# Extract flags
-if echo "${INPUT}" | grep -qE 'github\.com/.*/issues/[0-9]+'; then
+AUTO_YES=false
-  echo "GitHub URL detected - fetching issue..."
+PRIORITY=3
-  gh issue view "${INPUT}" --json number,title,body,labels,url
+LABELS=""
-else
+
-  echo "Text input detected - analyzing clarity..."
+# Parse using parameter expansion
-  # Continue with text parsing
+while [[ $INPUT == -* ]]; do
-fi
+  case $INPUT in
+    -y|--yes)
+      AUTO_YES=true
+      INPUT="${INPUT#* }"  # Remove flag and space
+      ;;
+    --priority)
+      PRIORITY="${INPUT#* }"
+      PRIORITY="${PRIORITY%% *}"  # Extract next word
+      INPUT="${INPUT#*--priority $PRIORITY }"
+      ;;
+    --labels)
+      LABELS="${INPUT#* }"
+      LABELS="${LABELS%% --*}"  # Extract until next flag
+      INPUT="${INPUT#*--labels $LABELS }"
+      ;;
+    *)
+      INPUT="${INPUT#* }"
+      ;;
+  esac
+done
+
+# Remaining text is the main input (GitHub URL or description)
+MAIN_INPUT="$INPUT"
 ```
 
-Then follow the workflow based on detected input type.
+### Execution Flow (All Phases)
+
+```
+1. Parse Arguments (Phase 0)
+   └─ Extract: AUTO_YES, PRIORITY, LABELS, MAIN_INPUT
+
+2. Detect Input Type & Clarity (Phase 1)
+   ├─ GitHub URL/Short? → Score 3 (clear)
+   ├─ Structured text?   → Score 2 (somewhat clear)
+   ├─ Long text?        → Score 1 (vague)
+   └─ Short text?       → Score 0 (very vague)
+
+3. Extract Issue Data (Phase 2)
+   ├─ If GitHub: gh CLI fetch + parse
+   └─ If text: Parse structure + apply PRIORITY/LABELS flags
+
+4. Context Hint (Phase 3, conditional)
+   └─ Only for clarity 1-2 AND no components → ACE search (max 3 files)
+
+5. Clarification (Phase 4, conditional)
+   └─ If clarity < 2 AND NOT auto mode → Ask for details
+   └─ If auto mode (AUTO_YES=true) → Skip, use inferred data
+
+6. GitHub Publishing (Phase 5, conditional)
+   ├─ If source = github → Skip (already from GitHub)
+   └─ If source != github:
+      ├─ If auto mode → Default NO (keep local)
+      └─ If manual → Ask user preference
+
+7. Create Issue (Phase 6)
+   ├─ Create local issue via ccw CLI
+   └─ If publishToGitHub → gh issue create → link
+
+8. Output Result (Phase 7)
+   └─ Display: ID, title, source, GitHub status, next step
+```
+
+## Quick Examples
+
+```bash
+# Auto mode - GitHub issue (no questions)
+/prompts:issue-new -y https://github.com/org/repo/issues/42
+
+# Standard mode - text with priority
+/prompts:issue-new "Database connection timeout" --priority 1
+
+# Auto mode - text with priority and labels
+/prompts:issue-new --yes "Add caching layer" --priority 2 --labels "enhancement,performance"
+
+# GitHub short format
+/prompts:issue-new GH-123
+
+# Complex text description
+/prompts:issue-new "User login fails. Expected: redirect to dashboard. Actual: 500 error"
+```

.codex/prompts/lite-execute.md

@@ -1,6 +1,6 @@
 ---
-description: Execute tasks based on in-memory plan, prompt description, or file content (Codex Subagent Version)
+description: Execute tasks based on in-memory plan, prompt description, or file content with optimized Codex subagent orchestration. Supports multiple input modes and execution control.
-argument-hint: "[--in-memory] [\"task description\"|file-path]"
+argument-hint: "[--plan=in-memory|<file-path>] [--parallel] [--skip-tests] [--dry-run]"
 ---
 
 # Workflow Lite-Execute Command (Codex Subagent Version)
@@ -28,14 +28,15 @@ Flexible task execution command with **optimized Codex subagent orchestration**.
 ### Command Syntax
 ```bash
 /workflow:lite-execute [FLAGS] <INPUT>
-
-# Flags
---in-memory                Use plan from memory (called by lite-plan)
-
-# Arguments
-<input>                    Task description string, or path to file (required)
 ```
 
+### Flags
+
+- `--plan=in-memory|<file-path>`: Input mode (in-memory plan or file path)
+- `--parallel`: Execute tasks in parallel (default: sequential)
+- `--skip-tests`: Skip test execution
+- `--dry-run`: Preview execution without making changes
+
 ## Input Modes
 
 ### Mode 1: In-Memory Plan

.codex/prompts/lite-fix.md

@@ -1,6 +1,6 @@
 ---
-description: Lightweight bug diagnosis and fix workflow with optimized Codex subagent patterns (merged mode)
+description: Lightweight bug diagnosis and fix workflow with optimized Codex subagent patterns. Supports severity and scope control.
-argument-hint: BUG="<bug description or error message>" [HOTFIX="true"]
+argument-hint: "BUG=\"<description or error message>\" [--hotfix] [--severity=critical|high|medium|low] [--scope=<path>]"
 ---
 
 # Workflow Lite-Fix Command (Codex Optimized Version)
@@ -28,6 +28,10 @@ Intelligent lightweight bug fixing command with **optimized subagent orchestrati
 **Target bug**: $BUG
 **Hotfix mode**: $HOTFIX
 
+- `--hotfix`: Hotfix mode, prioritize speed
+- `--severity`: Bug severity (critical|high|medium|low)
+- `--scope`: Debug scope limit (file path)
+
 ## Execution Modes
 
 ### Mode Selection Based on Severity

.codex/prompts/lite-plan-a.md

@@ -1,6 +1,6 @@
 ---
-description: Lightweight interactive planning workflow with single-agent merged mode for explore → clarify → plan full flow
+description: Lightweight interactive planning workflow with single-agent merged mode for explore → clarify → plan full flow. Supports depth control and auto-clarification.
-argument-hint: TASK="<task description or file.md path>"
+argument-hint: "TASK=\"<task description or file.md path>\" [--depth=standard|deep] [--auto-clarify] [--max-rounds=<n>] [--verbose]"
 ---
 
 # Workflow Lite-Plan-A (Merged Mode)
@@ -35,6 +35,10 @@ Single-agent merged mode for lightweight planning. One agent handles exploration
 
 **Target task**: $TASK
 
+- `--depth`: Exploration depth (standard|deep)
+- `--auto-clarify`: Auto clarify, skip confirmation
+- `--max-rounds`: Max interaction rounds
+
 ## Execution Process
 
 ```

.codex/prompts/lite-plan-b.md

@@ -1,6 +1,6 @@
 ---
-description: Lightweight interactive planning workflow with hybrid mode - multi-agent parallel exploration + primary agent merge/clarify/plan
+description: Lightweight interactive planning workflow with hybrid mode - multi-agent parallel exploration + primary agent merge/clarify/plan. Supports agent count and iteration control.
-argument-hint: TASK="<task description or file.md path>"
+argument-hint: "TASK=\"<task description or file.md path>\" [--num-agents=<n>] [--max-iterations=<n>] [--angles=role1,role2,...]"
 ---
 
 # Workflow Lite-Plan-B (Hybrid Mode)
@@ -35,6 +35,10 @@ Hybrid mode for complex planning tasks. Multiple agents explore in parallel from
 
 **Target task**: $TASK
 
+- `--num-agents`: Number of parallel agents (default: 4)
+- `--max-iterations`: Max iteration rounds
+- `--angles`: Exploration angles (role1,role2,...)
+
 ## Execution Process
 
 ```

.codex/prompts/lite-plan-c.md

@@ -1,6 +1,6 @@
 ---
-description: Lightweight interactive planning workflow with Codex subagent orchestration, outputs plan.json after user confirmation
+description: Lightweight interactive planning workflow with Codex subagent orchestration, outputs plan.json after user confirmation. Supports depth and exploration control.
-argument-hint: TASK="<task description or file.md path>" [EXPLORE="true"]
+argument-hint: "TASK=\"<description or file.md path>\" [--depth=standard|deep] [--explore] [--auto]"
 ---
 
 # Workflow Lite-Plan Command (Codex Subagent Version)
@@ -22,6 +22,10 @@ Intelligent lightweight planning command with dynamic workflow adaptation based
 **Target task**: $TASK
 **Force exploration**: $EXPLORE
 
+- `--depth`: Exploration depth (standard|deep)
+- `--explore`: Force exploration phase
+- `--auto`: Auto mode, skip confirmation
+
 ## Execution Process
 
 ```