How do I call an API in Python?

Install requests once (pip install requests), then a call is three lines that matter: make the request, check it worked, read the JSON.

import requests

response = requests.get(
    "https://api.github.com/repos/python/cpython",
    timeout=10,
)
response.raise_for_status()          # stop here on 404, 500 and friends
repo = response.json()               # the body, parsed into a dict

print(repo["full_name"], "has", repo["stargazers_count"], "stars")

That block needs your own machine (new to running Python there? start with the terminal): browsers only let a page call servers that explicitly allow it, so it can't run here. Everything after response.json() is ordinary Python, and you can run that part. Here is the same handling on a dict shaped like GitHub's reply (the numbers are made up):

data = {
    "full_name": "python/cpython",
    "stargazers_count": 68000,
    "topics": ["python", "interpreter"],
    "license": {"name": "Other"},
}

print(data["full_name"], "has", data["stargazers_count"], "stars")
print("topics:", ", ".join(data.get("topics", [])))
print("licence:", data.get("license", {}).get("name", "none"))

python/cpython has 68000 stars
topics: python, interpreter
licence: Other

The chained .get(..., {}) is worth copying: real APIs leave fields out, and one missing key shouldn't crash the whole run.

The three lines people skip

• timeout=10. requests has no timeout by default. If the server accepts the connection and then says nothing, your script waits forever. Ten seconds is a sensible starting point.
• raise_for_status(). A 404 or 500 is not an exception in requests; it's a normal response with an error code. Without this line, .json() runs on an error page and you get a confusing JSONDecodeError instead of a clear "404 Not Found".
• params= instead of building the URL by hand. It encodes spaces and & correctly:

import requests

response = requests.get(
    "https://api.github.com/search/repositories",
    params={"q": "web scraping language:python", "per_page": 5},
    timeout=10,
)
response.raise_for_status()
for repo in response.json()["items"]:
    print(repo["full_name"])

Sending data

json= turns a dict into a JSON body and sets the header for you. Keep keys out of your code and read them from the environment:

import os
import requests

response = requests.post(
    "https://api.example.com/v1/orders",
    json={"product": "Notebook", "quantity": 2},
    headers={"Authorization": f"Bearer {os.environ['SHOP_API_KEY']}"},
    timeout=10,
)
response.raise_for_status()
print(response.json()["id"])

The APIs lesson carries on from here: pagination, retries with back-off, and rate limits.