runt console - nteract Desktop

The runt console command launches a standalone Jupyter kernel and provides an interactive Read-Eval-Print-Loop (REPL) in your terminal.

Usage

runt console [KERNEL] [OPTIONS]

Arguments

KERNEL

string

Name of the kernel to launch (e.g., python3, julia, deno).If omitted, you must provide a custom command with --cmd.

Options

cmd

string

Custom command to launch the kernel. Use {connection_file} as a placeholder for the connection file path.Example: --cmd "deno jupyter --kernel --conn {connection_file}"

verbose

boolean

Print all Jupyter protocol messages for debugging.

Examples

Python console

runt console python3

Output:

python3 console
Use Ctrl+D to exit.

In [1]: import numpy as np

In [2]: np.array([1, 2, 3]) * 2
Out[2]: array([2, 4, 6])

In [3]:

Custom kernel command

runt console --cmd "deno jupyter --kernel --conn {connection_file}"

Debug mode (verbose)

runt console python3 --verbose

Shows all Jupyter messages:

python3 console
Use Ctrl+D to exit.

In [1]: 1 + 1
[iopub] status (ours=true)
[iopub] execute_input (ours=true)
[iopub] execute_result (ours=true)
Out[1]: 2
[iopub] status (ours=true)
[shell] execute_reply (ours=true)

Features

Input/Output Handling

Standard output: Streamed to terminal as it’s produced
Standard error: Displayed in red
Return values: Shown as Out[N]: ...
Errors: Full traceback displayed

Stdin Support

The console supports input() calls from the kernel:

In [1]: name = input("Enter your name: ")
Enter your name: Alice

In [2]: print(f"Hello, {name}")
Hello, Alice

Password prompts are hidden:

In [1]: import getpass

In [2]: pw = getpass.getpass("Password: ")
Password: [hidden]

Execution Count

Each input/output is numbered sequentially, matching the Jupyter notebook experience:

In [1]: x = 5
In [2]: x * 2
Out[2]: 10

Keyboard Shortcuts

Key	Action
`Ctrl+D`	Exit the console
`Ctrl+C`	Interrupt running code (sent to kernel)

Exit Behavior

When you exit the console (Ctrl+D or EOF):

A graceful shutdown request is sent to the kernel
The kernel process is terminated
Connection files are cleaned up

In [5]: [Ctrl+D pressed]

Shutting down kernel...
Done.

Advanced Usage

Launch Julia kernel

runt console julia

Launch R kernel

runt console ir

Custom Deno kernel

runt console --cmd "deno jupyter --kernel --conn {connection_file}"

Connection Files

The console creates a kernel connection file at:

~/.local/share/jupyter/runtime/runt-kernel-{id}.json

This file is automatically cleaned up on exit.

Troubleshooting

Kernel not found

Error: Kernelspec 'python3' not foundSolution:

List available kernels: jupyter kernelspec list

Install the kernel:

# Python kernel
pip install ipykernel
python -m ipykernel install --user

# Deno kernel
deno jupyter --install

Kernel hangs or doesn't respond

Symptoms: Commands don’t execute, no output appearsSolution:

Press Ctrl+C to interrupt the kernel
Exit with Ctrl+D and restart the console
Check for errors: runt console python3 --verbose

Output display issues

Symptoms: Rich output (plots, HTML) doesn’t displayNote: The console only displays plain text output. Rich media types (images, HTML, LaTeX) are ignored. Use runt notebook for full output rendering.

runt jupyter ps — List running kernels
runt jupyter stop — Stop a kernel
runt jupyter exec — Execute code in a running kernel
runt notebook — Launch full notebook UI with rich output

Commands

​Usage

​Arguments

​Options

​Examples

​Python console

​Custom kernel command

​Debug mode (verbose)

​Features

​Input/Output Handling

​Stdin Support

​Execution Count

​Keyboard Shortcuts

​Exit Behavior

​Advanced Usage

​Launch Julia kernel

​Launch R kernel

​Custom Deno kernel

​Connection Files

​Troubleshooting

​Related Commands

Build docs developers (and LLMs) love