OpenAI Launches Symphony to Automate Engineering Workflows Directly From Task Trackers

Apr 27, 2026 · Updated May 6, 2026

OpenAI launched Symphony, an open-source agent orchestrator (a coordination layer managing multiple AI agents) for the Codex platform. It integrates with task trackers like Linear to identify issues and trigger autonomous "implementation runs." These runs allow agents to navigate codebases and execute tests in isolated environments without constant human supervision.

This release provides the orchestration layer for the Codex multi-agent command center. While previous updates introduced durable execution and secure sandboxes, Symphony connects these capabilities to the source of engineering work. It transforms development into a continuous pipeline mirroring industrial-scale engineering automation.

You can use Symphony to turn a static backlog into an active work queue where agents handle routine bug fixes. The framework is open source, allowing for custom logic and integration with various project management tools. It is available now on GitHub for developers building production-grade agentic workflows.

View the full update on openai.com

OpenAI Developers

@OpenAIDevsApr 27

📣 What if every open issue had a Codex agent? That’s the idea behind Symphony, an open-source agent orchestrator for Codex that turns task trackers into always-on systems for agentic work, letting humans focus on review and direction. https://t.co/TxPs0bdtRd

851.4k

View on X

Still wondering? A few quick answers below.

Symphony is an open-source agent orchestrator created by OpenAI to manage Codex agents. It connects directly to task trackers to transform them into always-on systems for autonomous work. This framework allows AI agents to pick up and complete engineering tasks independently, shifting the human role from manual coding to high-level review and direction.

Symphony functions by polling project management tools, such as Linear, to find open issues that are ready for development. When a task is identified, the orchestrator initiates an autonomous implementation run. During this process, a Codex agent works through the necessary steps to resolve the issue in an isolated environment, ensuring the work is handled systematically.

Yes, Symphony is an open-source framework. OpenAI released it to provide developers with a customizable orchestration layer for building agentic engineering workflows. Being open source allows teams to modify the system logic, integrate it with different tools, and maintain control over how autonomous agents interact with their specific codebases and internal project management systems.

An implementation run is a multi-step autonomous process where a Codex agent attempts to solve a specific task or bug. Symphony manages these runs in isolated environments, allowing the agent to edit code and execute tests without interfering with the main codebase. This structured approach ensures that agentic work is contained, verifiable, and ready for human review.

Symphony is primarily designed for developers and engineering teams who want to implement agentic workflows at scale. It is particularly useful for organizations looking to automate routine development tasks and bug fixes. By providing an orchestration layer, it helps technical leads manage autonomous agents as team members rather than just using them as reactive chat assistants.