Menu
Skills How it works Pricing
Members Video skillGraphic & Type15–60 sec

Halftone Dossier (Feed)

A 9:16 case-file Skill for “Why X is secretly Y” lists and countdowns, with cream halftone paper, misregistered headlines, slamming stamps and numbered exhibits.

Theme A: a cat case file. Theme B: a walking-versus-gym list. Two reviewed 23-second renders. 9:16 · 0:23

What you get

  • 1080×1920 / 30 fps MP4 (15–60 s), cover.jpg, captions.srt and layout.json
  • Halftone print paper, misregistered slab headlines, slamming rubber stamps and giant numbered exhibits
  • A replaceable protagonist: a drawn critter, a drawn object mascot, or your own transparent cut-out with a print treatment
  • Local Kokoro narration with burned captions; synthesised stamp, shutter and wipe sounds (no music shipped)
  • Scene-by-scene editing: change one line and only that scene re-renders

You will need

  • Node.js 22+, FFmpeg and FFprobe on macOS
  • HyperFrames 0.8.181 (fetched by npx on first render)
  • Optional local Kokoro voice: Python with kokoro-onnx and soundfile (or render with --no-voice)
  • A list topic with a hook line, 3–5 short counts and a verdict

Good fit

  • 'Why your X is secretly Y' and '3 reasons' list videos
  • Countdowns, top-N roundups and number reveals
  • Playful myth-busting, product comparisons and habit explainers
  • Pet, lifestyle and consumer-tech accounts that want a recognisable print look

Not for

  • Talking-head footage, live-action b-roll or photo slideshows
  • Serious news, medical or legal claims presented as fact
  • Long-form explainers past 60 seconds or 16:9 masters
  • Lip-synced characters or photographic protagonists that need to act

You say

“Use Halftone Dossier (Feed) to turn a 'why X is secretly Y' list into a vertical case-file video. Write the brief (hook, suspect, numbered counts, verdict, closing line), confirm it, then create the project and run node render.mjs. Edit by scene id; only changed scenes re-render.”

After installing, say something like this to your agent with your own topic and assets. It asks only for what is missing.

FAQ

Do I need an account?

Yes. Members Skills install from a signed-in FeedSkills account with an active Founding Membership.

What does it cost to run?

The Skill is included with Founding Membership. It needs the tools listed under “You will need”. Any voice, model or media service you choose bills you directly; FeedSkills does not include provider credits.

Can I use the videos commercially?

Videos you make are yours to use, including commercially. The Skill package is licensed as: Source available · upstream MIT preserved. Fonts, music, footage and services you add keep their own terms.

Which agents does it work with?

Agents that read SKILL.md files, such as Codex or Claude Code. What FeedSkills has actually tested is listed under Tested scope below.

Tested scope

Version
0.1.2
Acceptance
This version has a recorded acceptance entry.
Tested
Two 23-second 1080×1920 / 30 fps example films (a cat 'case file' in the classic palette and a walking-vs-gym list in the manila palette) rendered with HyperFrames 0.8.181, local Kokoro narration (bm_george), burned captions plus captions.srt and layout.json; a one-line edit re-rendered only the edited scene (frame hashes of the other six scenes unchanged). Feed checks: 0 fail, 0 warn.
Runs on
Claude Code · Codex · macOS · Node.js 22+ · FFmpeg
License
Source available · upstream MIT preserved

Details

A case file for fast list videos

Halftone Dossier (Feed) turns a short list into a vertical case file: an accusation hook, a suspect profile, numbered exhibits, a verdict stamp and a colour-crash close. It keeps the original print character—cream paper, halftone overprint, misregistered headlines and rough rubber stamps—while adapting the composition for a readable 9:16 feed.

This release has passed FeedSkills’ visual and rhythm review. The public preview shows two real 23-second renders: a classic-palette cat case file and a manila-palette walking-versus-gym list. They are examples of the Skill, not stock footage or a promise of performance.

Two reviewed themes

What you get

  • A 1080×1920 / 30 fps MP4 workflow for 15–60 second Shorts, Reels and TikToks.
  • A replaceable critter, sneaker or transparent image subject with a print treatment.
  • A hook, profile, count, verdict and close scene grammar with local Kokoro narration, burned captions and captions.srt.
  • layout.json and a render report so the safe area, text size, contrast and scene reuse can be checked.
  • Two editable English brief examples: examples/theme-a.json and examples/theme-b.json.

Use it with your agent

Halftone Dossier (Feed) is included with Founding Membership. The public registry exposes only this summary; the Skill source and its ZIP are not public downloads. Sign in and use the install prompt above after your membership is active.

Create each project outside the installed Skill directory. Start with a topic, protagonist, hook, three to five counts and a closing action. Review the full video with sound before publishing; a successful render or machine check does not replace that review.

Runtime and boundaries

Rendering runs locally with Node.js 22+, FFmpeg/FFprobe and HyperFrames 0.8.181. Optional Kokoro narration downloads model data on first use. No music, personal voice, provider credentials or third-party stock media are bundled. Use subjects and source material you have permission to publish; a photo cut-out cannot blink or perform expressions.

The Skill is for playful lists, countdowns, explainers and comparisons. It is not for talking-head footage, photo slideshows, serious news, medical or legal claims presented as fact, or long-form and horizontal masters.

License and boundaries

The style and engine adapt LemoLab’s Halftone Dossier at pinned commit 413a6bd6d78ebd3bdd5abaec9538a26867f890df. Its code remains under the upstream MIT license; no upstream film, character, story, captions, music or stills are distributed. FeedSkills’ 9:16 scene grammar, palettes, narration, captions, replaceable subjects and scene-level editing are the adaptation in this package.