Codex MCP Server
Advanced MCP server wrapper for OpenAI Codex CLI v0.50.0+ that provides enterprise-grade conversational AI coding assistance with session management, intelligent model selection, and native resume functionality.
Prerequisites
OpenAI Codex CLI v0.50.0+ must be pre-installed and configured
Install:
npm i -g @openai/codexorbrew install codexSetup: Run
codex login --api-key "your-openai-api-key"⚠️ Breaking Change:
OPENAI_API_KEYenvironment variable is no longer supported⚠️ Version Requirement: v0.50.0+ required (see Version Compatibility below)
Claude Code installed
Version Compatibility
This MCP server requires codex CLI v0.50.0 or later due to the following changes:
v0.50.0+: Introduced
--skip-git-repo-checkflag (now required)v0.50.0+: Removed
--reasoning-effortflag (no longer supported)
If you have an older version of codex CLI, you will need to upgrade:
npm update -g @openai/codexFor detailed version compatibility information, see docs/codex-cli-integration.md.
Installation
One-Click Installation
VS Code
VS Code Insiders
Cursor
Manual Installation
Claude Code
claude mcp add codex-cli -- npx -y codex-mcp-serverClaude Desktop
Add to your Claude Desktop configuration file:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%/Claude/claude_desktop_config.json
{
"mcpServers": {
"codex-cli": {
"command": "npx",
"args": ["-y", "codex-mcp-server"]
}
}
}Usage in Claude Code
Once installed, Claude Code can use these tools:
codex - AI Coding Assistant
codex - AI Coding AssistantAsk Codex to analyze code, generate solutions, or provide coding assistance with optional session support for conversational context.
Basic Usage:
Use the codex tool to explain this function:
[paste your code here]Advanced Usage:
# Model selection (defaults to gpt-5-codex)
Use codex with model "gpt-4" to analyze this complex algorithm
# Reasoning effort control
Use codex with reasoningEffort "high" for thorough code review
# Session with model preferences (using advanced GPT-5 Codex by default)
Use codex with sessionId "my-session" and model "gpt-4" to refactor this code
# Continue conversation (uses native codex resume)
Use codex with sessionId "my-session" to make it more efficient
# Reset session context
Use codex with sessionId "my-session" and resetSession true to start fresh analysisParameters:
prompt(required): Your coding question or requestsessionId(optional): Session ID for conversational contextresetSession(optional): Reset session history before processingmodel(optional): Specify model to use (defaults to 'gpt-5-codex'). Options: 'gpt-5-codex', 'gpt-4', 'gpt-3.5-turbo'reasoningEffort(optional): Control reasoning depth ('low', 'medium', 'high')
listSessions - Session Management
listSessions - Session ManagementList all active conversation sessions with metadata including creation time, last access, and turn count.
Usage:
Use listSessions to see all active coding sessionsping - Connection Test
ping - Connection TestTest if the MCP server is working properly.
help - Codex CLI Help
help - Codex CLI HelpGet information about Codex CLI capabilities and commands.
Example Workflows
Code Analysis:
Please use the codex tool to review this TypeScript function and suggest improvementsConversational Code Development:
# Start a session for complex refactoring
Use codex with sessionId "refactor-auth" to analyze this authentication system
# Continue building on the analysis
Use codex with sessionId "refactor-auth" to implement the security improvements you suggested
# Check session history
Use listSessions to see all active development sessionsBug Fixing:
Use codex to help debug this error: [error message]Code Generation:
Ask codex to create a React component that handles file uploadsAdvanced Features (Codex CLI v0.50.0+)
Session Management
Key Features:
Native Resume: Leverages
codex resumefor optimal conversation continuityIntelligent Defaults: GPT-5-Codex model for superior coding assistance
Fallback Context: Manual context building when resume unavailable
Session Persistence: Context maintained across interactions (24hr TTL)
Production-Ready: Comprehensive error handling and graceful degradation
Enterprise-Scale: 54 tests covering all functionality and edge cases
Model Selection & Control
Key Features:
Default GPT-5 Codex: Automatically uses the latest
gpt-5-codexmodel for optimal coding assistanceDynamic Models: Choose between different models per request
Reasoning Effort: Control AI processing depth (low/medium/high)
Per-Model Optimization: Optimized prompts for different model families
Flexible Configuration: Mix and match model and session parameters
Enhanced Authentication
Requirements:
⚠️ Breaking Change: Environment variable
OPENAI_API_KEYno longer supportedNew Setup: Run
codex login --api-key "your-key"for authenticationCredential Storage: Credentials stored securely in
CODEX_HOME/auth.json
Best Practices:
Use sessions with native resume for complex, multi-step development tasks
Default Model: Uses
gpt-5-codexby default for optimal coding assistanceChoose appropriate reasoning effort:
lowfor quick answers,highfor complex analysisSelect models based on task complexity:
gpt-5-codexfor coding (default),gpt-4for advanced reasoning,gpt-3.5-turbofor speedReset sessions when switching to unrelated topics
Check
listSessionsto manage active conversations
Development
# Install dependencies
npm install
# Development mode
npm run dev
# Build
npm run build
# Start built server
npm startDocumentation
📚 Comprehensive Guides
Session Management - Advanced session features and implementation details
Codex CLI Integration - v0.50.0+ features, breaking changes, and migration guide
API Reference - Complete tool documentation and usage examples
🔧 Development Resources
Testing Suite: 54 comprehensive tests covering all functionality
TypeScript Support: Full type definitions and IntelliSense
Error Handling: Robust error recovery and graceful degradation
Performance: Optimized for enterprise-scale usage
License
ISC
Last updated
Was this helpful?