An open-source spec for orchestration: Symphony

2026-04-09 · Source: OpenAI News · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering · Depth: Advanced, extended

Summary

OpenAI has released Symphony, an open-source agent orchestrator designed to transform project management boards like Linear into control planes for coding agents. Developed to overcome context-switching bottlenecks encountered with interactive coding agents, Symphony assigns a dedicated agent to each open task, enabling continuous operation and human review. This system has reportedly led to a 500% increase in landed pull requests for some teams at OpenAI. Symphony decouples work from individual coding sessions and pull requests, allowing agents to manage larger units of work, create follow-up tasks, and shepherd changes through CI pipelines. The core of Symphony is a `SPEC.md` file, defining the problem and intended solution, rather than a complex supervision system, allowing agents high-level steering.

Key takeaway

For CTOs and VPs of Engineering aiming to scale developer productivity and reduce context switching, adopting an agent orchestration system like Symphony offers a compelling path. By integrating coding agents directly with your issue tracker, you can automate routine development tasks, accelerate feature delivery, and free your engineers to focus on more complex, high-value problems. Consider implementing Symphony to transform your project management into an always-on, agent-driven workflow, potentially achieving significant increases in output and operational efficiency.

Key insights

Symphony orchestrates coding agents via issue trackers, boosting productivity by automating task management and reducing human oversight.

Principles

• Decouple work from sessions and PRs.
• Treat issue trackers as control planes.
• Delegate routine tasks to agents.

Method

Symphony continuously monitors an issue tracker (e.g., Linear), assigns a dedicated agent to each active task, and manages its lifecycle, including restarts, task creation, and CI pipeline navigation, based on a `WORKFLOW.md` specification.

In practice

• Use `WORKFLOW.md` to define agent prompts and runtime settings.
• Configure `tracker.active_states` and `terminal_states` for task flow.
• Implement `hooks` for workspace lifecycle events.

Topics

• Agent Orchestration
• Symphony Specification
• Coding Agents
• Project Management Integration
• Automated Software Development

Code references

• openai/symphony

Best for: CTO, VP of Engineering/Data, Director of AI/ML, Software Engineer, AI Engineer, AI Architect

Related on AIssential

• OpenAI Open-Sources Symphony, a SPEC.md for Autonomous Coding Agent Orchestration — Symphony orchestrates autonomous coding agents using project management tools to overcome human attention bottlenecks.
• Conductors to Orchestrators: The Future of Agentic Coding — Software engineering is shifting from direct coding to orchestrating autonomous AI agent fleets for scaled productivity.
• OpenAI Releases Symphony: An Open Source Agentic Framework for Orchestrating Autonomous AI Agents through Structured, Scalable Implementation Runs — Symphony automates AI-assisted coding via structured, verifiable agentic workflows managed by an Elixir-based framework.
• OpenAI Symphony vs Claude Managed Agents vs CrewAI: Which Agent Orchestration Pattern Wins — Research has already identified effective AI agent orchestration patterns before recent product releases.
• Agent Bricks Supervisor Agent is Now GA: Orchestrate Enterprise Agents — Databricks' Supervisor Agent centralizes enterprise AI agent orchestration with built-in governance and continuous learning.
• AgentCraft: Putting the Orc in Orchestration — Ido Salomon — Effective multi-agent orchestration requires enhanced visibility, delegated autonomy, and seamless human-agent collaboration.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by OpenAI News.