Skip to content
Prism
Use cases

Social media videos in After Effects with AI

Yes, you can make social media videos in After Effects with AI: ChatGPT, Claude or any MCP client connected to Prism builds the hook, captions, cards and end card as editable layers in your open comp, in every format.

What ranks for this is template packs and browser editors: pick a preset, export an MP4, leave After Effects. A Reel is a hook, captions, a few cards and an end card, and each is a layer you describe.

What you say, and what lands

You typeWhat appears in your comp
"Hook at 0 s: 'Stop scrolling' fills the top third, slams in with overshoot, and 'this took four minutes' types on underneath."Two text layers, scale keyframes with overshoot on the first, a write-on animator on the second
"Transcribe the clip and caption it three words at a time, Anton, all caps, low centre, the spoken word in yellow."One text layer per group on the spoken timestamps, a text animator that colours the current word
"Card at 6 s: rounded rectangle, 'Step 2 of 3', slides in from the right, holds two seconds, slides out."A shape layer and a label, position keyframes in and out with a named ease
"End card for the last three seconds: logo centred, handle underneath, 'Follow for part 2', everything fades up."The logo from your project, two text layers, opacity keyframes on the last three seconds

Test it on your own project. A Free account is enough to start: Sign up, then connect your AI and ask for this.

How to do it

1
Open a comp with the clip in it
1080×1920 for Reels, Shorts and TikTok. Prism acts on the open comp, so the footage goes in first.
2
Name the beats
Hook at the start, card at six seconds, end card on the last three.
3
Ask for the captions
Transcription is a Prism AI capability, paid from your balance. It marks every word, and the caption layers are built from those marks. See word by word subtitles.
4
Describe the style once
Font, colours, safe margins. "Keep everything above the bottom 320 px" is a rule the layout follows.
5
Review it in the timeline
Every hook, caption and card is a native layer you can retime or restyle.
6
Ask for the other formats
"Rebuild this as 1080×1350 and 1920×1080" makes the other comps, each laid out for its frame. See multiple aspect ratios.

One edit, every format

FrameSizeWhere it goesWhat usually changes
9:161080×1920Reels, Shorts, TikTokthe master
4:51080×1350Instagram feedcaptions rise, hook shrinks
1:11080×1080feed, LinkedIncards move to the centre
16:91920×1080YouTubecaptions go wider and lower

Repurpose an existing video

You haveYou askWhat lands
a 16:9 talking-head clip"Put it in a 9:16 comp, scale to fill, centre the face, then caption it."The footage scaled and positioned, captions on the words
a finished 16:9 explainer comp"Rebuild this for 9:16 with the text stacked."A new comp, every layer re-laid, keyframes kept

You pick the moment

Prism does not pick the sixty seconds for you. You give the in and out; the captions and cards are built for that stretch.

What it needs

After Effects2024, 2025 or 2026, macOS or Windows
AI clientany MCP client: ChatGPT, Claude, Claude Code, Cursor, Codex and others
Plana Free account is enough for the layout; transcription is a Prism AI capability, paid from your balance

What it does not do

It does not trim your footage into highlights, record a voiceover, pull clips from a stock library or post to any platform. There is no template pack: the style is the one you describe, reused by asking for it again. Every change is undoable in After Effects as usual.

Common questions