Hip-hop captions, in the page's voice.
A ready-to-paste caption system for hip-hop and internet-culture clip pages. It reads the post's own text — the original caption plus any on-screen text (OCR) in the image — and writes with contractions and asymmetric beats in a fixed four-beat structure: react, ask, engage, tags. It never sounds like a tool wrote it.
Four beats, every time.
React, ask, engage, tags — the structure that gets replies and shares without sounding like a brand.
Asymmetric rhythm
A short beat, then a longer one — the way people actually type.
Reads the picture
On-screen text (OCR) grounds the react beat when the post has no caption of its own.
At most one punchline
One quoted fragment, never a wall of them.
Lowercase + contractions
Ain't, gotta, tryna — the voice, not a summary.
From file to on-voice in four steps.
No code, no API keys — just a .md you paste into the agent you already use.
The .md caption pack.
You're buying a downloadable .md pack written for AI agents (and humans). It carries the complete caption system — system prompt, inputs, output contract, and voice knobs — plus the universal anti-slop rules.
Chinchilla Caption Agent Pack vs. ad-hoc prompting.
Most people retype the same caption instructions into a chat box every single day, and the voice drifts each time. This pack is a system: eight structured prompt sets with input contracts, output contracts, and the anti-slop rules that keep a page sounding human.
Keep the Universal rules. The pack opens with four cross-cutting rules — never disclose the pipeline, never add unearned CTAs, concrete beats hype, match the page's real voice — so every category inherits the same discipline.
Get the Hip-Hop Caption Pack.
Free download. Unlock caption-hiphop-pack.md — paste the prompt into your agent and it writes in the page's voice from the first run.
25 units), capped at 60s