From model to agent: Equipping the Responses API with a computer environment
📰 OpenAI News
OpenAI's Responses API is equipped with a computer environment to execute real-world tasks, enabling agents to handle complex workflows beyond model capabilities
Action Steps
- Use the Responses API with the shell tool to propose and execute actions
- Utilize the hosted container workspace for isolated execution with filesystem and network access
- Implement reusable skills and context compaction for more efficient workflows
Who Needs to Know This
Developers and engineers on a team can benefit from this technology to build more sophisticated and automated workflows, while product managers can leverage it to create more efficient and scalable products
Key Insight
💡 Equipping models with a computer environment enables them to handle complex workflows and execute real-world tasks
Share This
🤖 OpenAI's Responses API gets a computer environment boost! 🚀
Key Takeaways
OpenAI's Responses API is equipped with a computer environment to execute real-world tasks, enabling agents to handle complex workflows beyond model capabilities
Full Article
# From model to agent: Equipping the Responses API with a computer environment | OpenAI
[Skip to main content](https://openai.com/index/equip-responses-api-computer-environment#main)
[](https://openai.com/)
* [Research](https://openai.com/research/index/)
* Products
* [Business](https://openai.com/business/)
* [Developers](https://openai.com/api/)
* [Company](https://openai.com/about/)
* [Foundation(opens in a new window)](https://openaifoundation.org/)
Log in[Try ChatGPT(opens in a new window)](https://chatgpt.com/)
* Research
* Products
* Business
* Developers
* Company
* [Foundation(opens in a new window)](https://openaifoundation.org/)
[Try ChatGPT(opens in a new window)](https://chatgpt.com/)Login
OpenAI
Table of contents
* [The shell tool](https://openai.com/index/equip-responses-api-computer-environment#the-shell-tool)
* [Orchestrating the agent loop](https://openai.com/index/equip-responses-api-computer-environment#orchestrating-the-agent-loop)
* [When the context window gets full: compaction](https://openai.com/index/equip-responses-api-computer-environment#when-the-context-window-gets-full-compaction)
* [Container context](https://openai.com/index/equip-responses-api-computer-environment#container-context)
* [Agent skills](https://openai.com/index/equip-responses-api-computer-environment#agent-skills)
* [How agents are made](https://openai.com/index/equip-responses-api-computer-environment#how-agents-are-made)
* [Make your own agent](https://openai.com/index/equip-responses-api-computer-environment#make-your-own-agent)
March 11, 2026
[Engineering](https://openai.com/news/engineering/)
# From model to agent: Equipping the Responses API with a computer environment
By Bo Xu, Danny Zhang, and Rohit Arunachalam
Loading…
Share
We're currently in a shift from using models, which excel at particular tasks, to using agents capable of handling complex workflows. By prompting models, you can only access trained intelligence. However, giving the model a computer environment can achieve a much wider range of use cases, like running services, requesting data from APIs, or generating more useful artifacts like spreadsheets or reports.
A few practical problems emerge when you try to build agents: where to put intermediate files, how to avoid pasting large tables into a prompt, how to give the workflow network access without creating a security headache, and how to handle timeouts and retries without building a workflow system yourself.
Instead of putting it on developers to build their own execution environments, we built the necessary components to equip the [Responses API(opens in a new window)](https://developers.openai.com/api/docs/guides/migrate-to-responses) with a computer environment to reliably execute real-world tasks.
OpenAI’s Responses API, together with the shell tool and a hosted container workspace, is designed to address these practical problems. The model proposes steps and commands; the platform runs them in an isolated environment with a filesystem for inputs and outputs, optional structured storage (like SQLite), and restricted network access.
In this post, we’ll break down how we built a computer environment for agents and share some early lessons on how to use it for faster, more repeatable, and safer production workflows.
## The shell tool
A good agent workflow starts with a tight execution loop: the model proposes an action like reading files or fetching data with API, the platform runs it, and the result feeds into the next step. We’ll start with the shell tool—the simplest way to see this loop in action—and then cover the container workspace, networking, reusable skills, and context compaction.
To understand the shell tool, it’s first useful to understand how a language model uses tools in general: to do things like call a function or interact with a computer. During training, a model is shown examples of how tools are used and the resulting effe
[Skip to main content](https://openai.com/index/equip-responses-api-computer-environment#main)
[](https://openai.com/)
* [Research](https://openai.com/research/index/)
* Products
* [Business](https://openai.com/business/)
* [Developers](https://openai.com/api/)
* [Company](https://openai.com/about/)
* [Foundation(opens in a new window)](https://openaifoundation.org/)
Log in[Try ChatGPT(opens in a new window)](https://chatgpt.com/)
* Research
* Products
* Business
* Developers
* Company
* [Foundation(opens in a new window)](https://openaifoundation.org/)
[Try ChatGPT(opens in a new window)](https://chatgpt.com/)Login
OpenAI
Table of contents
* [The shell tool](https://openai.com/index/equip-responses-api-computer-environment#the-shell-tool)
* [Orchestrating the agent loop](https://openai.com/index/equip-responses-api-computer-environment#orchestrating-the-agent-loop)
* [When the context window gets full: compaction](https://openai.com/index/equip-responses-api-computer-environment#when-the-context-window-gets-full-compaction)
* [Container context](https://openai.com/index/equip-responses-api-computer-environment#container-context)
* [Agent skills](https://openai.com/index/equip-responses-api-computer-environment#agent-skills)
* [How agents are made](https://openai.com/index/equip-responses-api-computer-environment#how-agents-are-made)
* [Make your own agent](https://openai.com/index/equip-responses-api-computer-environment#make-your-own-agent)
March 11, 2026
[Engineering](https://openai.com/news/engineering/)
# From model to agent: Equipping the Responses API with a computer environment
By Bo Xu, Danny Zhang, and Rohit Arunachalam
Loading…
Share
We're currently in a shift from using models, which excel at particular tasks, to using agents capable of handling complex workflows. By prompting models, you can only access trained intelligence. However, giving the model a computer environment can achieve a much wider range of use cases, like running services, requesting data from APIs, or generating more useful artifacts like spreadsheets or reports.
A few practical problems emerge when you try to build agents: where to put intermediate files, how to avoid pasting large tables into a prompt, how to give the workflow network access without creating a security headache, and how to handle timeouts and retries without building a workflow system yourself.
Instead of putting it on developers to build their own execution environments, we built the necessary components to equip the [Responses API(opens in a new window)](https://developers.openai.com/api/docs/guides/migrate-to-responses) with a computer environment to reliably execute real-world tasks.
OpenAI’s Responses API, together with the shell tool and a hosted container workspace, is designed to address these practical problems. The model proposes steps and commands; the platform runs them in an isolated environment with a filesystem for inputs and outputs, optional structured storage (like SQLite), and restricted network access.
In this post, we’ll break down how we built a computer environment for agents and share some early lessons on how to use it for faster, more repeatable, and safer production workflows.
## The shell tool
A good agent workflow starts with a tight execution loop: the model proposes an action like reading files or fetching data with API, the platform runs it, and the result feeds into the next step. We’ll start with the shell tool—the simplest way to see this loop in action—and then cover the container workspace, networking, reusable skills, and context compaction.
To understand the shell tool, it’s first useful to understand how a language model uses tools in general: to do things like call a function or interact with a computer. During training, a model is shown examples of how tools are used and the resulting effe
DeepCamp AI