n8n Social Post Gate & Rewrite with TypeSafe Jev + GPT-4o mini

Gate, score, and rewrite your n8n social posts with TypeSafe Jev + GPT-4o mini

Stop publishing guesswork. This n8n workflow drafts (or validates) social post copy, evaluates it with TypeSafe Jev for quality and overclaim risk, and automatically rewrites failing drafts with OpenAI (gpt-4o-mini). Posts that still don’t meet your thresholds are sent to Telegram for human review—so you can publish with confidence.

What this workflow does

• Accepts inputs when run manually or triggered by another workflow (e.g., product facts, audience, platform, and an optional existing draft).
• Generates a draft with OpenAI only when no draft is provided—creating an under-80-word post based on the supplied product facts.
• Scores the post with TypeSafe Jev for overall quality and flags issues such as overclaiming, weak hook, lack of clarity, hard-sell tone, typos, and off-topic content.
• Decides next action by comparing Jev results against your minimum quality and maximum overclaim thresholds.
• Rewrites when needed: OpenAI rewrites the copy using the Jev feedback, then the workflow re-scores until it passes or the rewrite limit is reached.
• Escalates to humans by sending the post (and scores) to Telegram if it still fails after maximum rewrites.

Use cases

• Automate SaaS social posting with guardrails to reduce compliance risk (overclaims) and improve clarity.
• Turn product facts from an internal source into platform-ready copy without manual editing.
• Run an editorial review pipeline: approve automatically or route uncertain drafts to Telegram.

Technical details

• OpenAI for draft + rewrite (default configured model: gpt-4o-mini).
• TypeSafe Jev via HTTP request using TypeSafe API key as HTTP Bearer Auth.
• Telegram integration to send unresolved posts for human review.
• Workflow logic uses n8n nodes including if, set, code, no op, switch, and telegram.