Scraping Video Content to WordPress Drafts
Budget: $30 – $250 USD
I keep a Google Sheet that holds a single column of direct YouTube links. Each new URL should trigger an automated flow that:
• pulls info from full transcript (or captions or subtitles, whatever) for the video,
• gives it to ChatGPT via API
• make ChatGPT also look on the web for info on the topic (from title column in the Google Sheets, but if that is not provided then use video's title)
• ChatGPT uses both the info from video and web and uses a custom prompt I will supply, and
• saves the resulting long-form article as a draft post—complete with title, slug, and any categories/tags I define—via the WordPress REST API.
Python is my natural first choice because of the rich ecosystem around Google Sheets API, YouTube Data API, BeautifulSoup/SerpAPI for web snippets, the OpenAI API, and wp-api, but I’m open to Node or another language if you can show equal reliability. Cron-friendly execution, clear logging, basic rate-limit handling, and a “processed” flag back to the Sheet are must-haves so the job can run hands-free.
Deliverables
1. Well-commented source code (ZIP). Avoid github.
2. README with setup steps, .env sample, and cron example
3. One-click deployment guide (Docker, Cloud Function, or similar)
4. A quick Loom or text walkthrough proving a sample video ends up as a draft in my WordPress dashboard
I’d like the first usable version as soon as reasonably possible; we can refine prompts and formatting together once the core loop is stable.
For full info and sample Google Sheets file, please check these links:
https://docs.google.com/document/d/11RzKArYFSzyoljKyN3-378sRmWjLDJnazOFVLuVuMtk/edit?tab=t.0
• pulls info from full transcript (or captions or subtitles, whatever) for the video,
• gives it to ChatGPT via API
• make ChatGPT also look on the web for info on the topic (from title column in the Google Sheets, but if that is not provided then use video's title)
• ChatGPT uses both the info from video and web and uses a custom prompt I will supply, and
• saves the resulting long-form article as a draft post—complete with title, slug, and any categories/tags I define—via the WordPress REST API.
Python is my natural first choice because of the rich ecosystem around Google Sheets API, YouTube Data API, BeautifulSoup/SerpAPI for web snippets, the OpenAI API, and wp-api, but I’m open to Node or another language if you can show equal reliability. Cron-friendly execution, clear logging, basic rate-limit handling, and a “processed” flag back to the Sheet are must-haves so the job can run hands-free.
Deliverables
1. Well-commented source code (ZIP). Avoid github.
2. README with setup steps, .env sample, and cron example
3. One-click deployment guide (Docker, Cloud Function, or similar)
4. A quick Loom or text walkthrough proving a sample video ends up as a draft in my WordPress dashboard
I’d like the first usable version as soon as reasonably possible; we can refine prompts and formatting together once the core loop is stable.
For full info and sample Google Sheets file, please check these links:
https://docs.google.com/document/d/11RzKArYFSzyoljKyN3-378sRmWjLDJnazOFVLuVuMtk/edit?tab=t.0
Related categories:
PHP
Python
WordPress
Web Scraping
Docker
Data Extraction
BeautifulSoup
API Integration
REST API
OpenAI