[Feature Request]: GPT-5.1 shell tool support

[priyansh4320]

Is your feature request related to a problem? Please describe.

The shell tool allows the model to interact with your local computer through a controlled command-line interface. The model proposes shell commands; your integration executes them and returns the outputs. This creates a simple plan-execute loop that lets models inspect the system, run utilities, and gather data until they can finish the task.

Shell is available through the Responses API for use with GPT-5.1. It is not available on other models, or via the Chat Completions API.

when to uses

Automating filesystem or process diagnostics — For example, “find the largest PDF under ~/Documents” or “show running gunicorn processes.”
Extending the model’s capabilities — Using built-in UNIX utilities, python runtime and other CLIs in your environment.
Running multi-step build and test flows — Chaining commands like pip install and pytest.
Complex agentic coding workflows — Using other tools like
apply_patch
to complete workflows that involve complex file operations.

Response includes shell_call output items

When the model decides to use the shell tool, it will include one or multiple shell_call output items. Each shell_call output item has the following fields.

call_id is generated by the API and should be passed back in shell_call_output in the next request.
status indicates whether the tool call is completed or in_progress (in streaming mode).
action contains information for shell tool execution
commands: A list of shell commands that developers can execute concurrently in the local computer environment.
timeout_ms: Optional timeout that should be honored in shell execution.
max_output_length: Used to truncate the output when it’s too large. Developer needs to pass it back with the raw output in shell_call_output, and does not need to do the actual truncation.

{
    "type": "shell_call",
    "call_id": "...",
    "action": {
        "commands": ["ls -l"],
        "timeout_ms": 120000,
        "max_output_length": 4096
    },
    "status": "in_progress"
}

sources :

https://platform.openai.com/docs/guides/tools-shell#use-shell-tool-with-responses-api

Describe the solution you'd like

No response

Additional context

No response