MCP
Turn one video into a week of content
Blog post, X thread, LinkedIn post, newsletter section, Shorts script — one upload, six assets, same day.
run_youtube_toolThe problem
You publish a video, and the treadmill immediately resets: the next one is due, so the last one never gets the distribution it earned. The blog version doesn't get written. The X thread stays a "later" note. The newsletter goes out without it.
Here's the thing — the video you already made contains the week of content. The argument is your blog post. The spikiest lines are your thread. The professional angle is your LinkedIn post. The story is your newsletter section. The best forty seconds are your Short. Repurposing isn't creative work, it's extraction work — which is exactly why it always loses to making the next video, and exactly why it shouldn't be done by hand.
How veridive MCP helps
run_youtube_tool is a catalog of 32 content tools that run on any YouTube video, straight from your AI assistant. Point it at your upload and ask for the derivative you need: blog-post, x-post, linkedin-post, newsletter, shorts-script, plus packaging tools like title-generator, description-generator and tag-extractor.
Every output is generated from your actual transcript — your argument, your examples, your phrasing, restructured for the destination format. That's the difference between repurposing and generic AI filler: nothing is invented, it's your material changing shape. You review, tweak the voice where it drifts, and ship.
It all runs inside Claude, ChatGPT or Cursor through one MCP connection — describe the output you want and your assistant picks the right tool from the catalog.
How to run it
- 1Connect veridive MCP
One-time setup: add the veridive MCP server to Claude, ChatGPT, Cursor or any MCP client and generate your API key. Takes about two minutes.
- 2Start with the anchor asset
Give your assistant the video link and ask for the first derivative. It runs the matching tool from the catalog against your transcript and returns a structured draft.
Take this video and turn it into a blog post that keeps my argument and my voice: <YouTube link>. Use veridive's run_youtube_tool. - 3Generate the rest of the batch
Same video, different shapes. Ask for each platform's format in one go — every draft comes from the same transcript, so the story stays consistent across channels.
Now from the same video: an X thread, a LinkedIn post, a newsletter section and a Shorts script. Make each one native to its platform, not a copy-paste of the blog post. - 4Package the next upload while you're at it
The catalog also covers packaging: generate title options, a description and tags for your next video — and run the hook grader on your opening to see how it scores before you publish.
For this video, give me three title options, a description and tags — and grade my opening hook: <YouTube link>
Problems it solves
- Repurposing is the first thing cut when time is short — so content you spent days on gets exactly one day of distribution.
- Rewriting the same idea five platform-native ways burns the creative energy you need for the next video.
- Generic AI writing loses your voice; drafts extracted from your own transcript keep your argument, examples and phrasing.
- A thread is not a blog post is not a newsletter — each format has its own craft, and doing all of them manually never survives a real schedule.
Why it's valuable
- One upload becomes five or six assets the same day — compounding distribution for work you already did.
- Every draft is grounded in your transcript, so editing is a polish pass, not a rewrite.
- It works on any public video — including podcasts you guested on: turn someone else's episode into your LinkedIn post.
- No new app in the workflow: it runs inside the assistant you already write with.
Example prompts
Turn this video into a blog post that keeps my argument and voice: <YouTube link> — use veridive's run_youtube_tool.From the same video, write an X thread and a LinkedIn post — each native to its platform.Give me three title options, a description and tags for this video, and grade my opening hook: <YouTube link>FAQ
Which of the 32 tools does this workflow use?
Mainly the writing tools — blog-post, x-post, linkedin-post, newsletter, shorts-script — plus the packaging tools: title-generator, description-generator, tag-extractor and hook-grader. You never call them by name unless you want to; describe the output and your assistant picks the tool.
Will the outputs actually sound like me?
They're built from your transcript — your argument, your examples, your phrasing — restructured for each format. Expect a strong draft that needs a voice pass, not a rewrite. The more distinct your spoken style, the more of it survives.
Does it work on videos that aren't mine?
Yes — any public YouTube video. The classic move: you guest on a podcast, then turn that episode into your own thread, post and newsletter section without asking the host for a transcript.
Do I have to name the tools in my prompts?
No. Your assistant picks the right veridive tool from a plain-language request — "turn this into a blog post" routes to run_youtube_tool's blog-post on its own. Naming helps only when you have several content servers connected and want the routing to be unambiguous.
Related use cases
Most-replayed segments plus comment sentiment show what actually hooked viewers — pattern-match it before your next upload.
Comment summarizer surfaces the questions your audience keeps asking under your videos and your rivals' — answered demand, pre-validated.
Fresh-upload filters find the week's real signal; the newsletter tool drafts the issue with your framing.
Run this in your own assistant
Connect the veridive MCP server to Claude, ChatGPT or Cursor and paste the prompts above — setup takes about two minutes.