Faceless YouTube Doodle Image Set from a Timestamped Transcript

Turn a timestamped script into consistent MS Paint-style image prompts, sortable timestamp filenames, and an ffmpeg draft-video file.

Creator / YouTubefaceless-youtubeimage-prompts
by AgentVoltv1.0.0
Free to sign up · every skill included with AgentVolt Pro

Before & after

Same prompt, same agent — the only difference is the skill.

Prompt“Create 30 MS-Paint style images from a 3-minute script.”
Beforegeneric AI agent
Creating lots of images for a video can be tedious if you're doing it manually. There are a few approaches — you could transcribe the script and then generate images, or use existing image tools. The challenge is getting them to match a consistent style and making sure they're labeled properly for video editing. You'd probably want to transcribe your script first, identify key moments, then generate one image per section. I could help you set up a workflow that automates the labeling based on timestamps. What format is your script in right now?
Afterwith Faceless YouTube Doodle Image Set from a Timestamped Transcript
FACELESS YOUTUBE IMAGE BATCH WORKFLOW

STEP 1: TRANSCRIPTION + TIMESTAMPS
Input: 3-min voiceover (180 seconds)
Tool: Turboscribe (or similar)
Output: Transcript with every 6-second interval marked (0.00, 0.06, 0.12... 3.00)
Result: 30 timestamp anchors

STEP 2: IMAGE PROMPT GENERATION
For each 6-second block, extract 1-2 key concepts
Example timestamps:
0.00-0.06: "why spreadsheets fail teams"
0.06-0.12: "hidden costs of manual data entry"
0.12-0.18: "collaboration breaks down"
Generate 30 MS-Paint-style prompts (simple shapes, muted colors, minimalist)

STEP 3: IMAGE GENERATION BATCH
Prompt template: "MS-Paint doodle style, hand-drawn feel, pastel colors, simple shapes, no text, 800x600px"
Generate all 30 via Hicksfield skill in Claude

STEP 4: BATCH DOWNLOAD + FILENAME MAPPING
Each image auto-names with its timestamp (0.00.png, 0.06.png, 0.12.png)
Download to /video-assets/images/

STEP 5: VIDEO EDITING
Drag all 30 images into timeline; Premiere reads filenames and auto-syncs to voiceover based on timestamp sort

TIME SAVING: Manual workflow = 3 hours; this workflow = 20 minutes (batch generation).

About this skill

Paste a timestamped transcript and get every image your faceless video needs, in one consistent crude-doodle style, named so the edit assembles itself.

What it does

  • Parses bracketed, clock, and SRT timestamps and computes how long each image holds
  • Locks one style prompt so 100 images look like one video
  • Picks one concrete visual per line and flags abstract lines for review
  • Names files by index and timestamp so they sort and import in order

What you get

A style lock, a shot table with prompts and filenames, and an ffmpeg concat file plus command that renders a timed draft with your voiceover.

Best for

Faceless YouTube creators making stick-figure or doodle-style explainers who want visuals done in one pass.

SKILL.md preview

SKILL.md
---
name: faceless-youtube-doodle-images
description: Turn a timestamped voiceover transcript into a full set of crude, MS Paint-style doodle images for a faceless YouTube video - one locked style prompt, one visual idea per timestamp, sortable timestamp filenames, and a ready-made ffmpeg slideshow file so editing is a drag-and-drop pass. Use when the user says "make images for every timestamp in my script", "MS Paint style visuals for my faceless channel", "generate doodle images from this transcript", "one image every 7 seconds", "name the images by timestamp", or "turn my SRT into a shot list".
---

# Faceless YouTube Doodle Image Set from a Timestamped Transcript

… (sign up to view the full skill)
Sign up to view, copy, and install the full skill

More creator skills

View all Creator skills →