How to Generate YouTube Titles and Descriptions with ChatGPT's API

Titles decide whether a video gets clicked, and writing ten good options by hand is tedious. This guide uses the OpenAI API to return five title candidates plus a full description block from your transcript or summary, so you pick rather than invent.

What you need

• An OpenAI API key from platform.openai.com
• Python 3.9+ with the openai package
• A short summary or transcript of the video

Step 1: Install and authenticate

Step 2: Ask for structured JSON

Request the result as JSON so your script can parse it directly into an upload step instead of you copying and pasting. The response_format flag keeps the model from wrapping prose around the data.

import os, json
from openai import OpenAI

client = OpenAI(api_key=os.environ["OPENAI_API_KEY"])

PROMPT = (
    "From the video summary, return JSON with keys: "
    "titles (5 options, under 60 chars, no clickbait lies), "
    "description (2 short paragraphs), "
    "tags (10 lowercase keywords)."
)

def metadata(summary):
    res = client.chat.completions.create(
        model="gpt-5",
        response_format={"type": "json_object"},
        messages=[
            {"role": "system", "content": PROMPT},
            {"role": "user", "content": summary},
        ],
    )
    return json.loads(res.choices[0].message.content)

if __name__ == "__main__":
    data = metadata("A beginner walkthrough of soldering a through-hole kit.")
    print(json.dumps(data, indent=2))

Step 3: Run and review

Result

You get a parsed object with five titles, a description, and tags. Combine this with the upload guide to set metadata at upload time instead of in the web UI.