Hi everyone ![]()
I’ve built a scenario that does the following:
-
Lists all posts from my Facebook Page
-
Lists all user comments under those posts
-
Sends each comment’s text to ChatGPT (OpenAI module) to generate a reply
-
Posts the reply as a new comment on the post (not as a “reply to comment”)
It’s mostly working fine, but I want to improve it with a few key features:
Only reply once per comment, even if the scenario runs again later.
Save replied comment IDs in a Data Store (“Replied Comments”) to skip duplicates.
Reduce credit usage — currently it lists all posts and comments each time.
Eventually scale to multiple Facebook Pages (e.g. for different brands we manage).
I’d like to confirm:
-
The best practice for the Data Store + Filter setup so it doesn’t re-reply.
-
The correct logic to ensure ChatGPT’s response posts directly on the post (not nested as a reply).
-
Any optimisation tips for large pages with many posts/comments.
If anyone has an example blueprint, screenshots, or step-by-step advice for this “Facebook Comments + ChatGPT + Data Store” workflow, I’d really appreciate it ![]()
Thank you!