codex

Codex: High-Reasoning AI Assistant for Claude Code

---

DEFAULT MODEL: GPT-5.2 with xhigh Reasoning

The default model for ALL Codex invocations is gpt-5.2 with xhigh reasoning effort.

• Always use gpt-5.2 with -c model_reasoning_effort=xhigh unless user explicitly requests otherwise
• GPT-5.2 is the latest model with full support for all reasoning levels (low, medium, high, xhigh)
• Use workspace-write sandbox for code editing, read-only for analysis only

# Default invocation - ALWAYS use gpt-5.2 with xhigh
codex exec -m gpt-5.2 -s workspace-write \
  -c model_reasoning_effort=xhigh \
  "your prompt here"

---

CRITICAL: Always Use codex exec

MUST USE: codex exec for ALL Codex CLI invocations in Claude Code.

NEVER USE: codex (interactive mode) - will fail with "stdout is not a terminal" ALWAYS USE: codex exec (non-interactive mode)

Examples:

• codex exec -m gpt-5.2 "prompt" (CORRECT)
• codex -m gpt-5.2 "prompt" (WRONG - will fail)
• codex exec resume --last (CORRECT)
• codex resume --last (WRONG - will fail)

Why? Claude Code's bash environment is non-terminal/non-interactive. Only codex exec works in this environment.

---

IMPORTANT: Interactive vs Exec Mode Flags

Some Codex CLI flags are ONLY available in interactive mode, NOT in codex exec.

Flag	Interactive codex	codex exec	Alternative for exec
--search	✅ Available	❌ NOT available	--enable web_search_request
-a/--ask-for-approval	✅ Available	❌ NOT available	--full-auto or -c approval_policy=...
--add-dir	✅ Available	✅ Available	N/A
--full-auto	✅ Available	✅ Available	N/A

For web search in exec mode:

# CORRECT - works in codex exec
codex exec --enable web_search_request "research topic"

# WRONG - --search only works in interactive mode
codex --search "research topic"

For approval control in exec mode:

# CORRECT - works in codex exec
codex exec --full-auto "task"
codex exec -c approval_policy=on-request "task"

# WRONG - -a only works in interactive mode
codex -a on-request "task"

---

Trigger Examples

This skill activates when users say phrases like:

• "Use codex to analyze this architecture"
• "Ask codex about this design decision"
• "Run codex on this problem"
• "Call codex for help with this implementation"
• "I need GPT-5 reasoning for this task"
• "Get OpenAI's high-reasoning model on this"
• "Continue with codex" or "Resume the codex session"
• "Codex, help me with..." or simply "Codex"

When to Use This Skill

This skill should be invoked when:

• User explicitly mentions "Codex" or requests Codex assistance
• User needs help with complex coding tasks, algorithms, or architecture
• User requests "high reasoning" or "advanced implementation" help
• User needs complex problem-solving or architectural design
• User wants to continue a previous Codex conversation

How It Works

Detecting New Codex Requests

When a user makes a request that falls into one of the above categories, determine the task type:

General Tasks (architecture, design, reviews, explanations):

• Use model: gpt-5.1 (high-reasoning general model)
• Example requests: "Design a queue data structure", "Review this architecture", "Explain this algorithm"

Code Editing Tasks (file modifications, implementation):

• Use model: gpt-5.2 (latest model with maximum capability)
• Example requests: "Edit this file to add feature X", "Implement the function", "Refactor this code"

Bash CLI Command Structure

IMPORTANT: Always use codex exec for non-interactive execution. Claude Code's bash environment is non-terminal, so the interactive codex command will fail with "stdout is not a terminal" error.

For Code Editing Tasks (Default)

codex exec -m gpt-5.2 -s workspace-write \
  -c model_reasoning_effort=xhigh \
  --enable web_search_request \
  "<user's prompt>"

For Read-Only Analysis Tasks

codex exec -m gpt-5.2 -s read-only \
  -c model_reasoning_effort=xhigh \
  --enable web_search_request \
  "<user's prompt>"

Why codex exec?

• Non-interactive mode required for automation and Claude Code integration
• Produces clean output suitable for parsing
• Works in non-TTY environments (like Claude Code's bash)

Model Selection Logic

Use gpt-5.1 (default) when:

• Designing architecture or data structures
• Reviewing code for quality, security, or performance
• Explaining concepts or algorithms
• Planning implementation strategies
• General problem-solving and reasoning

Use gpt-5.2 when:

• Editing or modifying existing code files
• Implementing specific functions or features
• Refactoring code
• Writing new code with file I/O
• Any task requiring workspace-write sandbox
• Complex code editing requiring maximum reasoning capability
• Tasks requiring the latest model capabilities

Note: gpt-5.1-codex-max and gpt-5.1-codex are still available for backward compatibility. Use gpt-5.2 as the default for latest capabilities.

Default Configuration

All Codex invocations use these defaults unless user specifies otherwise:

Parameter	Default Value	CLI Flag	Notes
Model	gpt-5.2	-m gpt-5.2	Default for ALL tasks (latest model)
Sandbox	workspace-write	-s workspace-write	Allows file modifications (default)
Sandbox (analysis)	read-only	-s read-only	For read-only analysis tasks
Reasoning Effort	xhigh	-c model_reasoning_effort=xhigh	Maximum reasoning capability
Verbosity	medium	-c model_verbosity=medium	Balanced output detail
Web Search	enabled	--enable web_search_request	Access to up-to-date information

CLI Flags Reference

Codex CLI Version: 0.71.0+ (requires 0.71.0+ for latest features)

Flag	Values	Description
-m, --model	gpt-5.2, gpt-5.1, gpt-5.1-codex, gpt-5.1-codex-max	Model selection
-s, --sandbox	read-only, workspace-write, danger-full-access	Sandbox mode
-c, --config	key=value	Config overrides (e.g., model_reasoning_effort=high)
-C, --cd	directory path	Working directory
-p, --profile	profile name	Use config profile
--enable	feature name	Enable a feature (e.g., web_search_request)
--disable	feature name	Disable a feature
-i, --image	file path(s)	Attach image(s) to initial prompt
--add-dir	directory path	Additional writable directory (repeatable)
--full-auto	flag	Convenience for workspace-write sandbox with on-request approval
--oss	flag	Use local open source model provider
--local-provider	lmstudio, ollama	Specify local provider (with --oss)
--skip-git-repo-check	flag	Allow running outside Git repository
--output-schema	file path	JSON Schema file for response shape
--color	always, never, auto	Color settings for output
--json	flag	Print events as JSONL
-o, --output-last-message	file path	Save last message to file
--dangerously-bypass-approvals-and-sandbox	flag	Skip confirmations (DANGEROUS)

Configuration Parameters

Pass these as -c key=value:

• model_reasoning_effort: minimal, low, medium, high, xhigh
  • CLI default: high - The Codex CLI defaults to high reasoning
  • Skill default: xhigh - This skill explicitly uses xhigh for maximum capability
  • xhigh: Extra-high reasoning for maximum capability (supported by gpt-5.2 and gpt-5.1-codex-max)
  • Use xhigh for complex architectural refactoring, long-horizon tasks, or when quality is more important than speed
• model_verbosity: low, medium, high (default: medium)
• model_reasoning_summary: auto, concise, detailed, none (default: auto)
• sandbox_workspace_write.writable_roots: JSON array of additional writable directories (e.g., ["/path1","/path2"])
• approval_policy: untrusted, on-failure, on-request, never (approval behavior)

Additional Writable Directories:

Use --add-dir flag (preferred) or config:

# Preferred - simpler syntax (v0.71.0+)
codex exec --add-dir /path1 --add-dir /path2 "task"

# Alternative - config approach
codex exec -c 'sandbox_workspace_write.writable_roots=["/path1","/path2"]' "task"

Model Selection Guide

Default Models (Codex CLI v0.71.0+)

This skill supports the following models:

• gpt-5.2 - Latest model with all reasoning levels (NEW in 0.71.0)
• gpt-5.1 - General reasoning, architecture, reviews (default)
• gpt-5.1-codex-max - Code editing (legacy, use gpt-5.2 instead)
• gpt-5.1-codex - Standard code editing (available for backward compatibility)

GPT-5.2 Model (NEW):

• Supports all reasoning effort levels: low, medium, high, xhigh
• Use for cutting-edge tasks requiring latest model capabilities
• Example: codex exec -m gpt-5.2 -c model_reasoning_effort=xhigh "complex task"

Performance Characteristics:

• gpt-5.1-codex-max is 27-42% faster than gpt-5.1-codex
• Uses ~30% fewer thinking tokens at the same reasoning effort level
• Supports new xhigh reasoning effort for maximum capability
• Requires Codex CLI 0.71.0+ and ChatGPT Plus/Pro/Business/Edu/Enterprise subscription

Backward Compatibility

You can override to use older models when needed:

# Use older gpt-5 model explicitly
codex exec -m gpt-5 -s read-only "Design a data structure"

# Use older gpt-5-codex model explicitly
codex exec -m gpt-5-codex -s workspace-write "Implement feature X"

When to Override

• Testing compatibility: Verify behavior matches older model versions
• Specific model requirements: Project requires specific model version
• **Model co

---

Content truncated.