Debug a Python script with AI

What does this setup do?

Debug a Python script with AI: paste the traceback and the code, get a plain-words explanation of the cause and one fix to test.

Who is it for?

Hobbyists and non-programmers with a Python script that broke and an error message they cannot read.

How do you set it up?

Tools

• ChatGPT or Claude (free tier is fine)
• The full error message (the traceback), copied completely
• The script, or at least the 20 lines around the failing line

Steps

1. Copy the entire traceback, not just the last line. The last line names the error, but the lines above show where it happened.
2. Learn the one skill that makes everything else work: read a traceback from bottom to top. The final line is the error message with the exception name; moving up shows the calls from most recent to least recent, each with filename, line number, and the code that ran.
3. Paste the traceback plus the surrounding code into the prompt. Ask for the cause in plain words first, then one fix.
4. Check the cause against the common list before accepting it. Typical culprits: NameError (undefined name), TypeError (wrong type for the operation, like 1 + "1"), ValueError (bad value, like int("a")), IndexError, KeyError, ZeroDivisionError, ImportError, AttributeError.
5. Apply exactly one fix and run the script again. If the error changes, you made progress; paste the new traceback and repeat.
6. If the AI’s fix does not work after two tries, ask it to explain what the code is trying to do line by line instead. Understanding beats guessing.
7. When it works, ask the AI to summarize what was wrong in one sentence and save it. Error patterns repeat, and your future self will thank you.

Help me debug my Python script. I am not a programmer, so explain in plain words.

The full traceback:
[paste the ENTIRE error output here]

The code around the failing line:
[paste about 20 lines, marking which line the traceback points to]

What the script is supposed to do: [e.g. "read a CSV of expenses and print monthly totals"]

Give me:
1. What the error means in one plain sentence.
2. The most likely cause, pointing at the specific line.
3. ONE fix to try. Just one.
4. How to check whether the fix worked.

Rules:
- Do not rewrite my whole script. One fix at a time.
- If you are not sure, say so and give me the two most likely causes.

I will test each fix before asking for the next step.

Which chat tools work with this prompt?

Paste the prompt into any of these assistants (free or paid tiers).

• ChatGPT
• Claude

What tips make it smoother?

• The bottom-to-top traceback habit pays off forever. Five minutes learning it saves hours on every future error.
• Change one thing per test run. Two simultaneous fixes and you will never know which one worked.
• Save every fixed error with its one-sentence cause. Most people hit the same five errors for years; a personal list turns each into a two-minute fix.

Where do the claims come from?

Summaries use public sources linked below, not independent verification. How setups are picked.

• Read Real Python’s write-up (cdn.realpython.com)

  Real Python: How to read a traceback: bottom to top, error line last, call entries with filename and line number.

• Read Elena Kosourova (Dataquest)’s write-up (dataquest.io)

  Elena Kosourova (Dataquest): Common built-in exceptions and try/except basics for handling them.