paper-trail-video

Produces animated cut-paper geopolitics YouTube episodes from script through render and upload.

Updated Sep 11, 2026
One-click install
npx skills add https://github.com/agusoler000/paper-trail-geo --skill paper-trail-video-agusoler000
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: paper-trail-video
Source: https://github.com/agusoler000/paper-trail-geo/tree/main/skills/paper-trail-video
Command: npx skills add https://github.com/agusoler000/paper-trail-geo --skill paper-trail-video-agusoler000

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Producing a 15-17 minute animated geopolitics video involves many fragile steps—scripting, voice generation, lip-sync alignment, choreography, rendering, mixing, and publishing—and this Skill encodes the entire Paper Trail channel pipeline so each episode is produced consistently at near-zero credit cost. ## Core Features & Use Cases - End-to-end episode pipeline: Covers interrogation and scriptwriting, ElevenLabs voice generation via fal.ai, whisper/fal timestamp alignment, plane-based choreography, frame checking, audio mixing, and final delivery. - Custom render engine: Uses a deterministic cut-paper animation engine (motor v4) with maps, rigs, props, HUD layers, word-by-word subtitles, and rhythm auditing (ritmo.py) against measured retention benchmarks. - Shorts production: Generates 3-5 vertical 1080x1920 shorts with exclusive scripts per episode, reusing the episode's art assets. - Use Case: When asked to produce a new episode on a geopolitical topic, the Skill runs demand validation, writes the script with primary sources, generates voice and choreography, renders the video, and prepares the upload package. ## Quick Start Ask the assistant to produce a new Paper Trail episode on a chosen geopolitics topic following the pipeline in this skill.

Frequently Asked Questions about paper-trail-video

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I produce an animated YouTube episode with this pipeline?

Start with demand validation using produccion/demanda.py, then run the closed interrogation for editorial stance, write the script with primary sources, generate voice via fal.ai ElevenLabs, build the choreography against timestamps, and render with the motor engine before final mixing and delivery.

How are vertical YouTube Shorts generated from an episode?

Shorts are produced alongside the main episode with 3-5 exclusive scripts per episode, reusing the episode's art assets rather than cutting the master footage. They render natively at 1080x1920 with the body filling the screen and a hook limited to 3.5 seconds.

What tools does the video render engine use?

The engine combines a custom Python motor (motor.py v4) for cut-paper animation with PIL and ffmpeg for rendering, whisper or fal timestamps for alignment, and ElevenLabs voices through fal.ai. Rhythm is audited with ritmo.py using ffmpeg and numpy metrics.

Why do rendered episodes look static or slow-paced?

The measured cause is push zoom distributed across the whole shot, making long shots slower, plus idle animations calibrated below perceptible thresholds. The fix is speed-based push, stronger idle motion, motivated cuts that change content, and higher frame occupancy verified with ritmo.py.

What are the cost limits for producing a video?

There is a hard cap of USD 4 per production covering the episode plus its shorts, treated as a ceiling rather than a budget to spend. Voice generation costs roughly USD 0.10 per 1,000 characters via fal.ai, and existing assets must be reused before generating new ones.