Run Codex tasks asynchronously in parallel while saving context tokens.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "codex-mcp-async" yet — see the docs or source repo.
Use codex-mcp-async to start 3 background Codex tasks at once: 1) fix failing unit tests in the payments module; 2) add missing TypeScript types to the user service; 3) refactor the logging module to reduce duplication. Filter intermediate thinking logs and return only a result summary, changed files, and next-step suggestions for each task.
Returns concise results for three parallel tasks, including fix summaries, touched files, and suggested validation steps.
Use codex-mcp-async to run a large background code review on the current repository: inspect API error handling, naming consistency, and potential security risks. Do not output detailed reasoning; keep only the final issue list, sorted by severity, with actionable fixes.
Produces a compressed review report with prioritized issues and fixes while avoiding heavy context usage.
Use codex-mcp-async to create two parallel background tasks: Task A designs and implements an initial version of a 'bulk CSV user import' feature; Task B writes tests for it and checks edge cases. Filter thinking logs and provide a final combined summary of the implementation, test coverage, and risks.
Delivers a combined implementation-and-testing summary for quick feasibility and quality assessment.
Run long background jobs asynchronously and poll results without tool-call timeouts.
Offload boilerplate, type generation, and summaries to OpenAI for faster development.
Let AI coding tools safely call Codex for development and automation tasks.
Call Codex inside Claude Code for reviews, second opinions, and delegated coding.
Enable Claude Code to perform coding tasks through the OpenAI Codex CLI.
Connect Claude Code Desktop to Codex, GPT-5, and Gemini APIs via MCP.