Skip to content
SkillScript-led1,015 tokens in use

Net-New Video Editor

Turns fresh talking-head recordings into reversible vertical drafts from a JSON edit plan, rendered with FFmpeg and Pillow with captions, a hook card and QA frames.

3,862 stars on the repository MIT Licenseericosiu/ai-marketing-skills · net-new-video-editor
npx skills add ericosiu/ai-marketing-skills -s net-new-video-editor
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing Net-New Video Editor costs

31tokens idle
1,015tokens in use
1,528tokens, up to

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Up to adds the files its instructions link. Counts are approximate, made offline with cl100k_base.

Model dependence Script-led

Script-ledMixedModel-led

Most of this skill is scripts, so it behaves much the same whichever model runs it.

This describes the skill's structure, how much of it is scripts. It does not say how well it works. Scripts: scripts/net_new_video_editor.py.

Touches

shelldetectednetworkdetectedwrites filesdetected

Declared tags come from the skill's frontmatter; detected ones from its scripts, its code blocks and the commands it names in inline code. This is detected, not audited.

How it scores

Quality

How well the skill is written, counted from its own files: its frontmatter, its steps, examples, stated output and constraints, its size and whether every file it ships is one it names.How we score it →
Good
  • Frontmatter8/15
  • Structure9/25
  • Examples12/15
  • Output15/15
  • Size15/15
  • Resources6/15

Security

Starts at 100 and loses points for each thing in the skill's files that a reader should look at first.How we score it →
Good
  • Runs shell commands−5

    does this on the machine, and the frontmatter does not say so

    SKILL.md, line 15
  • Writes files−8

    does this on the machine, and the frontmatter does not say so

    scripts/net_new_video_editor.py, line 56
  • Reaches the network−10

    does this on the machine, and the frontmatter does not say so

    scripts/net_new_video_editor.py, line 185

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Good
  • Development activity88
  • Maintenance44
  • Community56
  • Maturity28
  • Momentum40

Net-New Video Editor keeps creative choices in JSON and pixel work in one bundled Python renderer. The agent sets up a project folder, inspects the takes, writes a clean edit plan and optionally a more aggressive one, validates each plan and renders it.

The render trims paired audio and video, joins the segments, makes a center-safe 9:16 frame, draws captions and a hook card, normalizes loudness and exports H.264 and AAC video with three QA frames and a report.

What You Get

  • A project folder with raw, transcript, screens, assets, exports and qa directories and a brief file to fill in
  • An intake report that lists each take with duration, size and codecs
  • A clean edit plan and an optional aggressive plan that differs only by named retention edits
  • A 1080 by 1920 MP4 with burned captions and a hook card, loudness normalized
  • Three QA frames from the opening, middle and ending plus a pass or fail report

Common Use Cases

  • Making a first vertical cut of a talking-head recording for review
  • Testing a gentler and a tighter version of the same video
  • Checking a render with a dry run before running FFmpeg

What Makes It Unique

The skill separates judgment from rendering: the agent writes and validates a JSON plan, and a deterministic script does the cutting. It states completion as one of four states, from done to blocked, and it refuses to substitute another recording, brand or asset library. It will not publish, upload, delete originals or overwrite an approved master without current explicit approval.

Under The Hood

The skill ships one Python script that needs FFmpeg, ffprobe and Pillow on the machine. The agent runs it through the shell with init, inspect, validate, render and qa commands. The script probes media, builds caption and hook images, calls FFmpeg to render the video and writes JSON reports, QA images and the export inside the project folder, and refuses paths that leave that folder. It makes no network calls. The instructions also tell the agent to run the repository’s version check and telemetry helper scripts when they exist, which are opt-in.

Join founders building with open source

Opinionated takes, migration guides, cost-saving tips, and insights from the open source ecosystem.

Subscribe on SubstackJoin 750+ subscribers