Start the IDE:

nix-shell
code .

Then have a look at the output from a curl command.

%%bash
curl -v httpbin.org/get

{
  "args": {}, 
  "headers": {
    "Accept": "*/*", 
    "Host": "httpbin.org", 
    "User-Agent": "curl/7.68.0", 
    "X-Amzn-Trace-Id": "Root=1-61fa462e-2dae841071a7bd19231dab3e"
  }, 
  "origin": "86.19.94.164", 
  "url": "http://httpbin.org/get"
}


  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
  0     0    0     0    0     0      0      0 --:--:-- --:--:-- --:--:--     0*   Trying 35.171.190.227:80...
* TCP_NODELAY set
* Connected to httpbin.org (35.171.190.227) port 80 (#0)
> GET /get HTTP/1.1
> Host: httpbin.org
> User-Agent: curl/7.68.0
> Accept: */*
> 
* Mark bundle as not supporting multiuse
< HTTP/1.1 200 OK
< Date: Wed, 02 Feb 2022 08:51:58 GMT
< Content-Type: application/json
< Content-Length: 253
< Connection: keep-alive
< Server: gunicorn/19.9.0
< Access-Control-Allow-Origin: *
< Access-Control-Allow-Credentials: true
< 
{ [253 bytes data]
100   253  100   253    0     0    577      0 --:--:-- --:--:-- --:--:--   576
* Connection #0 to host httpbin.org left intact

You can set environment variables in a code block:

%env FOO bar

env: FOO=bar

and it shows up in subprocesses launched from othe code cells:

%%bash
env | grep FOO || true

FOO=bar

Here's a prolog for plotting pretty graphs:

import matplotlib.pyplot as plt
import numpy as np

and here's a graph:

print("setting up X")
x = np.linspace(0,5,1000)
print("step: constructing y-data")
y = np.sin(x)
print("step: display info about y")
plt.plot(x,y)
print("okay, I'm done now")

setting up X
step: constructing y-data
step: display info about y
okay, I'm done now

Github renders notebooks automatically in the browser, but not (for some reason) if it's a README. To convert this notebook to Markdown so Github can render it on the project page:

!jupyter nbconvert --to markdown README.ipynb

[NbConvertApp] Converting notebook README.ipynb to markdown
[NbConvertApp] Support files will be in README_files/
[NbConvertApp] Making directory README_files
[NbConvertApp] Writing 2553 bytes to README.md