📱 Auto-comment on Instagram posts with GPT-4o, Phantombuster, and SharePoint
⚡ 586 views · 📱 Social Media & Email Marketing
💡 Pro Tip — Heads up — if this workflow’s Instagram scraping breaks (it happens whenever Meta updates their frontend), ScraperNode is a community node that handles profiles, posts, reels, and comments without any HTML parsing.
Description
Who’s it for
Social-media teams and influencers who want to grow engagement by automatically posting short, relevant comments on new Instagram posts under rotating hashtags—while staying under platform limits.
How it works / What it does
- Schedule Trigger fires every 2 h at a specified minute.
- Select Cookie chooses a rotating Instagram session-cookie by time slice.
- Generate Random Hashtag (GPT-4o) outputs a realistic AI/BPA tag.
- Hashtag Agent (Phantombuster) scrapes up to
ENV_MAX_POSTS_PER_HASHTAGrecent posts. - Get Random Post picks one; Create Comment (GPT-4o) crafts a ≤150-character reply in the chosen language.
- Builds
instagram_post_to_comment.csv, uploads to SharePoint, then launches the Auto-comment Agent which posts the reply. - Post URL is logged to
instagram_posts_already_commented.csvto avoid duplicates. - Wait nodes throttle launches to ≈80 comments/day.
How to set up
- Add credentials: Phantombuster API, SharePoint OAuth2, OpenAI API key.
- In SharePoint › “Phantombuster” folder create:
•instagram_session_cookies.txt– one cookie per line.
•instagram_posts_already_commented.csvwith headerpostUrl. - Edit Set ENV Variables to set default language, prompt, and max posts per scrape.
- Activate the workflow; it will comment every 2 h.
Requirements
- n8n 1.33 +
- Phantombuster Growth plan (API access)
- OpenAI account (GPT-4o)
- Microsoft 365 SharePoint tenant
How to customize
- Change tone/length: tweak prompt in Create Comment.
- Comment more often: raise
ENV_MAX_POSTS_PER_HASHTAG, adjust cron, and agent limits. - Swap SharePoint for Drive/Dropbox by replacing storage nodes.
🔗 Nodes Used
Phantombuster, Schedule Trigger, AI Agent, OpenAI Chat Model, Convert to File, Extract from File
📥 Import
Download workflow.json and import into n8n:
Workflow menu → Import from File