$ loading_
Expose OpenAI Codex CLI tools through MCP for execution and job management.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "codex-mcp-server" yet — see the docs or source repo.
Use the codex-mcp-server exec tool to run the test command in the current project and return the output and status.
Returns the command output, status information, and related job results.
First call skill list to see available skills, then run an appropriate skill and report progress and results.
Returns the skill list, execution status for the selected skill, and a summary of the final result.
List current jobs, check the status of a specific job, and call kill if the job is stuck or no longer needed.
Returns the job list, the target job status, and the result of the termination action.
Developers can connect this tool in an MCP-compatible client and let AI use Codex CLI capabilities such as execution, review, and skill invocation as named tools. This reduces manual context switching to the terminal.
When Codex tasks need ongoing tracking, users can check status, poll for results, list jobs, and terminate jobs when needed. This fits automation flows that require execution monitoring.
Developers can integrate code review into MCP workflows through the exposed review tool. This allows review requests and results to stay within the same toolchain.
This tool wraps the OpenAI Codex CLI as an MCP server. It exposes eight Codex capabilities as named tools for use with pi or codex.
According to the description, it exposes eight tools: exec, review, skill list, skill run, status, poll, list jobs, and kill. These cover execution, review, skill usage, and job management.
The provided material does not include installation steps or prerequisites. For details, see the source repository.
Enable Claude Code to perform coding tasks through the OpenAI Codex CLI.
Connect IDEs or AI assistants to Codex CLI for safe automation and code analysis.
Use Codex in Claude Code for execution, debugging, and web search.
Delegate coding tasks to Codex via MCP with security and result checks.
Connects MCP clients to local Codex CLI for autonomous coding and thread history management.
Forward MCP tools to coding agents for session-based development tasks.