If you already produce a curated roundup for email — the five stories worth knowing this week, one line each — you've already done the hard part of a social content pipeline. The curation work, the actual research and judgment about what's worth including, is finished before the automation ever runs. What's left is reformatting, which is exactly the kind of task a node chain handles well.
The insight: repurposing beats re-researching
A workflow built to discover newsworthy stories from scratch needs a relevance filter, a recency check, and a fair amount of tuning before it reliably surfaces things worth posting — see curated vs. reactive news content strategy for that tradeoff. A workflow built to repurpose a roundup you already curate skips almost all of that, because someone already decided what's worth including when they wrote the newsletter. The automation's job shrinks down to: pull the already-curated content, reshape it to fit a social format, publish it. That's a meaningfully smaller and more reliable problem than discovery.
The node chain
Pull the roundup content. If your newsletter platform or CMS exposes any URL that returns your roundup's content in structured form — a JSON export, an RSS-style feed, an API endpoint — an HTTP Request node can call it directly. Many newsletter tools expose exactly this even without a formal developer API; check for an export or feed URL before assuming you need one.
Reformat, don't redraft. A Custom Agent node here is doing compression, not composition — your newsletter entry might run two or three sentences, but a carousel slide wants a headline and one supporting line. Prompt for that explicitly: From this newsletter roundup entry, extract a headline under 60 characters and a one-sentence summary under 120 characters — do not add new information or opinion beyond what the source entry states. That last clause matters: the model's job is trimming your already-good content, not embellishing it.
Bind into a recap template. An Apply Template node pointed at a saved "weekly roundup" carousel — one slide per story, or one slide per headline with the full list on a closing slide — binds each reformatted entry into its {placeholder} fields. Since this is a recurring, low-stakes format (a recap that's slightly imperfect costs very little), it's a reasonable candidate for eventually running on a timer once you've watched it handle a few real weekly runs — see the pre-launch checklist before making that call.
Post. Once the slides look right in a test run, a Post node publishes the finished carousel to Instagram.
Where this differs from newsjacking
This is deliberately not the same pattern as automated newsjacking, and that's the point. Newsjacking needs speed and a tight review step because it's reacting to something breaking in real time. A roundup repurposing workflow is running against content that's already been through your own editorial judgment once — the newsletter itself — so the review burden per run is genuinely lower. That doesn't mean skip review entirely, especially early on, but it does mean this is one of the more forgiving formats to eventually trust running unattended, precisely because someone already vetted the content before the automation ever touched it.
The actual time saved
The realistic win here isn't "no work" — it's turning a manual reformatting task (open the newsletter, rewrite five entries into slide copy, place them in a template, publish) into a workflow that does the reformatting and leaves you a finished draft to glance at. If your team already spends twenty minutes a week manually turning the roundup into a carousel, that's the twenty minutes this pipeline is built to remove — not the actual curation, which stays exactly as valuable, and exactly as human, as it always was.