$ loading_
Run restartable multi-agent coding workflows with live monitoring and bug-finding control.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "Codex Workflows MCP Server" yet — see the docs or source repo.
Start a multi-agent workflow for this Node.js API project: one agent reproduces the login endpoint error, one finds the root cause and proposes a fix, and another writes and runs regression tests. Show live progress in the terminal dashboard and allow restart from the latest checkpoint if any step fails.
A durable multi-agent run with live status, phase results, restart points, and final bug-fix and test conclusions.
Use a bug-finding template to scan this repository, focusing on concurrency, error handling, input validation, and edge cases. Assign multiple agents to different modules, then summarize high-risk issues with actionable fixes and priorities.
A module-by-module issue list with severity, root-cause analysis, fix recommendations, and priority.
Resume the interrupted frontend refactoring workflow from the latest successful checkpoint. Keep previous analysis, reassign the failed agent’s tasks, and continue the remaining tests and validation steps.
The original workflow resumes with updated run status and completes with full refactoring and validation results.
Control two local Codex CLI agents in parallel for secure development workflows.
Get technical advice, code reviews, and explanations via Codex CLI.
Discover and message local coding agents from within Codex sessions.
Manage, monitor, and resume long-running Codex Desktop tasks reliably.
Index local repositories for semantic search and structured code understanding.
Expose OpenAI Codex CLI tools through MCP for execution and job management.