AI Content Repurposing One recording, ten assets, within a day
One recording becomes show notes, a blog post, five posts, a thread, a newsletter section and a clip list — for you to edit.
- Installed for you · one-time$249
- Run for you$297/month
- Custom build · this process$3,000–5,000 one-time
- Starts whenNew file in a folder (Google Drive / Dropbox) or RSS
Covers up to eight episodes or recordings a month and the standard asset set. Video editing itself, additional languages, or a custom asset set are a custom build. Custom builds start at $2,000 site-wide; this band is for this process and its integrations.
AI Content Repurposing is an automation Agentmelt installs in Notion. The free template is a file you import yourself; the kit ($49) adds the setup guide and prompts. Installed for you costs $249 one-time and is live within two working days of access, with a recorded walkthrough and 30 days of fixes. Run for you costs $297/month, with setup and maintenance included. A custom build of this process is $3,000–5,000 one-time; custom builds start at $2,000 site-wide. No call is needed for any option. Written by Max Zeshut; last revised 2026-09-18.
Reference stack: Claude + n8n. Node names are on each step; the file is under the template.
Founder at Agentmelt
Reviews every build before it goes live; the workflow escalates to a person, never past one.
What is content repurposing?
AI content repurposing turns one recording — a podcast episode, a webinar, a founder interview — into every derivative asset a marketing team would otherwise write by hand: show notes with timestamps, a blog post that restructures the conversation into an article, five LinkedIn posts with different hooks, an X thread, a newsletter section, and a clip list with the six strongest 30–90-second segments for the editor. The workflow transcribes with speaker labels, extracts the structure and the quotable lines, drafts each asset in your voice guide and within platform limits, and puts everything on one review page. The owner edits and approves; approved posts go to the scheduler and the blog draft to the CMS. Weekly engagement per hook style feeds back into the prompts.
- We record an hour of good conversation a week and publish a link to it.
- Show notes are written on Friday afternoon by whoever is free.
- LinkedIn gets the same sentence as X, and both read like a summary.
- The editor asks which moments to clip and nobody has re-listened to the episode.
- A repurposing agency costs more per month than the podcast.
If two of these are yours, this is the process to look at first. Free audit — or read on.
What it looks like once it runs.
Content repurposing is the most-sold automation workflow because the input is already made: a podcast episode, a webinar, a founder interview. The workflow transcribes it, extracts the structure and the quotable moments, drafts every derivative asset in your voice — show notes, a blog post, five LinkedIn posts, an X thread, a newsletter section, clip timestamps for the editor — and puts them in a review board and the scheduler within 24 hours.
- Tuesday 9:00 A new recording lands in the folder. It is transcribed with speakers and timestamps.
- 9:20 The structure is extracted: segments, the three most quotable lines, people and resources mentioned.
- 9:40 Show notes, a blog post, five posts, a thread, a newsletter section and a clip list are drafted.
- 10:00 One review page per episode; you edit and mark each asset ready.
- On approval Posts are scheduled on your calendar; the blog post is published.
One hour of recorded content becomes one post, because turning it into ten takes a writer two days. Episodes pile up unpromoted; the newsletter goes out late; the social calendar is empty in the weeks between launches.
Every episode produces a complete set of assets the next morning, each following your format and voice, each reviewed in one place. Publishing cadence goes from 'when someone has time' to every week without adding headcount.
Trigger, then 8 steps.
Seven of the eight steps run without anyone. Step 06 waits for a person.
New file in a folder (Google Drive / Dropbox) or RSS
Transcribe
Audio to a speaker-labelled transcript with timestamps via Deepgram, AssemblyAI or Whisper; the transcript is stored next to the recording.
Extract structure
Segments with titles and timestamps, key claims, the three most quotable lines, named people and resources mentioned, and the episode's single main idea.
Draft the long-form
Show notes with timestamps and a blog post that restructures the conversation into an article — not a transcript summary — in your voice guide, with the resources linked.
Draft the short-form
Five LinkedIn posts (one per key idea, different hooks), an X thread, a newsletter section, and a pull-quote graphic brief; each within platform limits and your style rules.
Clip list for the editor
Timestamps for the six strongest 30–90-second segments with a suggested caption and hook, exported to the editor's tool or a sheet.
Review board
One page per episode with every asset, the transcript and the clip list; the owner edits and marks each asset approved.
Schedule approved assets
Approved posts go to Buffer, Hootsuite or the platform APIs on the cadence in the calendar rules; the blog post to the CMS as a draft; the newsletter section to the ESP draft.
Report performance back
Weekly: engagement per asset and per hook style, so the prompts learn which hooks work for your audience.
Every episode produces a complete set of assets the next morning, each following your format and voice, each reviewed in one place
Every episode produces a complete set of assets the next morning, each following your format and voice, each reviewed in one place.
One decision, at step 06. Seven steps ran without you.
Each pattern page explains it in plain words and lists every automation that uses it.
The template is built on n8n (how the n8n AI Agent node works); the same steps run on Make, Zapier or plain code. The tool is chosen to fit your accounts, you own the result, and the template comes in all three formats.
Data it touches
- Recording source (Drive, Dropbox, RSS, Riverside, Zoom)
- Transcription API (Deepgram, AssemblyAI, Whisper)
- Voice and format guide, past top-performing posts
- Scheduler (Buffer, Hootsuite) and CMS
- Platform analytics for the feedback loop
Guardrails
- Nothing publishes without approval on the review board.
- Quotes are verbatim from the transcript with timestamps; the agent cannot invent a quote.
- Claims extracted from the episode are attributed to the speaker, not stated as fact by the brand.
- Style rules (no emojis, no hashtags, sentence length) are enforced by structured output and checked before the board.
Why repurposing is the ideal first workflow
The input exists, the output formats are fixed, quality is easy to judge, and the person who reviews it is the person who benefits. There is no integration risk — a folder in, a Notion board out — and the value is visible the next morning. Most teams that run this workflow for a month never go back to writing derivative posts by hand, and the episode backlog gets promoted for the first time.
Voice is a prompt, but only if you write it down
The difference between generic AI posts and posts that sound like the founder is a voice guide: sentence length, vocabulary, what you never say, three examples of posts that worked and why. The workflow's prompts include that guide, and the weekly performance report tells you which hooks your audience responds to, so the guide improves. Teams without a written voice guide get one as part of setup — it is the most valuable hour of the engagement.
The blog post is not a summary
A transcript summary reads like a transcript summary and ranks for nothing. The long-form step restructures the conversation into an article with a thesis, sections and the resources mentioned, so it can stand alone in search. That is the asset with the longest shelf life, and it is worth the separate prompt.
What it connects to.
| Tool | Role in this workflow |
|---|---|
| n8n | Pipeline, board, scheduling, reporting |
| Deepgram / AssemblyAI | Transcription with speakers and timestamps |
| Claude | Extraction and all drafts, in your voice |
| Notion | Review board |
| Buffer / Hootsuite / CMS | Publishing |
Want this running without building it?
$297/month
We set it up in your tools, run it on our servershost it on our n8n cloud (or deploy it to your own n8n, Make or Zapier), watch it, and fix it when something changes. Setup included, cancel monthly; the automation stays yours if you leave.
$3,000–5,000 one-time
Your systems, your rules, your edge cases. A one-off build delivered into your accounts to keepon Claude and n8n, delivered into your accounts to keep, with documentation and a recorded walkthrough.
Covers up to eight episodes or recordings a month and the standard asset set. Video editing itself, additional languages, or a custom asset set are a custom build.
Also sold one-time, installed in your own accounts, as part of the marketers package and the content creators package and the copywriters package and the graphic designers package and the marketing teams package — do it yourself $49 · installed for you $249 · done for you $490.
When this becomes a custom build.
The subscription covers the standard version — $297/month, setup and maintenance included. Three things move it to a custom build ($3,000–5,000 one-time), delivered into your accounts to keep. If one of them is you, say so on the audit and the plan prices both.
Video editing — cutting the clips, not only listing them.
Assets in several languages.
A custom asset set — a course module, a press release, a paid-ad variant — or above eight recordings a month.
Running the standard version first is the usual path: the exceptions it logs in the first month are the scope of the custom build, with your own numbers. How custom builds work.
Frequently asked questions.
Does it edit video clips?
It produces the clip list with timestamps, hooks and captions for your editor or a tool like Descript or Opus. Automated clip rendering is a custom addition.
Can it match our tone?
Yes — the voice guide is part of setup and the prompts follow it. The first two episodes are reviewed together to calibrate; after that the drafts need light edits.
Which schedulers and CMSs are supported?
Buffer, Hootsuite, Later and the X/LinkedIn APIs for social; WordPress, Webflow, Ghost and HubSpot for the blog; Mailchimp, Beehiiv, ConvertKit and HubSpot for the newsletter.
How is AI content repurposing different from a summary?
A summary condenses; repurposing restructures. The blog post is written as an article with its own argument, drawn from the transcript but not tracing it; each LinkedIn post takes one idea and one hook; the thread has its own order. The voice guide — your phrases, your banned words, examples of posts that worked — is what keeps ten assets from sounding like one machine.
Do we still need a content repurposing agency?
For the drafting and the scheduling, no; an agency’s edge is judgement and editing, which is exactly the review step a person keeps here. Many teams keep an editor for the clips and the final pass and drop the retainer for the drafts. Video editing itself — cutting the clips, captions, graphics — is separate: the workflow lists the moments, the cutting is a service or a custom build.
Which recording sources and schedulers work?
Recordings from a shared folder, a podcast RSS feed, or Riverside and Zoom exports; transcription through Deepgram, AssemblyAI or Whisper; scheduling through Buffer, Hootsuite or the platforms directly; blog drafts into WordPress, Webflow, Ghost or HubSpot. Review happens in Notion or a similar board.
What this blueprint is built on.
Vendor documentation the steps rely on, and the rules the guardrails are written against.
- Deepgram — documentation developers.deepgram.com/home
- AssemblyAI — documentation www.assemblyai.com/docs
- n8n docs — Wait node (approval steps) docs.n8n.io/integrations/builtin/core-nodes/n8n-nodes-base.wait/