メニュー
OCR extracted storyboard frames, write a text manifest, and surface evidence-backed UI text from a local video.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Assemble a grounded review prompt from a local artifact bundle so Codex or Claude Code can inspect the right files and ask the right questions.
Check a local video against layout annotations, safe zones, OCR text boxes, and frame samples so agents can catch cropped content, caption overlap, unreadable overlays, and platform-unsafe composition before shipping.
Collect first-pass source media facts, audio silence/energy, shot estimates, representative frame status, and persistent text/caption risk evidence.
Review a rendered video for technical defects such as resolution mismatch, missing or near-silent audio, black or white frames, edge gutters, low motion, sparse caption bands, and layout issue pass-through.
Fuse video shot segments with storyboard frames, OCR, transitions, and timeline artifacts into one per-segment evidence map for grounded video review.
Extract at least one storyboard frame per video shot segment so later OCR and segment evidence fusion can cover gaps missed by global sampling.
| name | storyboard-ocr |
| description | OCR extracted storyboard frames, write a text manifest, and surface evidence-backed UI text from a local video. |
Use this after storyboard-extract when an agent needs machine-readable
UI text from extracted frames instead of only images.
This skill owns:
storyboard.ocr.jsonProvide either the storyboard directory or manifest path:
{
"storyboardDir": "./output/storyboard",
"minConfidence": 45
}
{
"manifestPath": "./output/storyboard/storyboard.manifest.json",
"minConfidence": 55
}
minConfidence defaults to 45 and must be between 0 and 100.
The skill writes storyboard.ocr.json beside the storyboard manifest. The
manifest includes per-frame OCR lines, semantic UI-like lines, confidence,
regions, optional bounding boxes, frame quality, unique text summaries, and
sampling metadata carried forward from extraction.
Run after storyboard-extract. Run before storyboard-transitions and
storyboard-understand. If OCR quality is weak but the visual frames look
important, manually inspect the images instead of relying only on text output.
Use semanticLines for product/UI claims; raw lines may include subtitles,
decorative text, or OCR noise. quality.status: reject means downstream text
inference should be low-trust for that frame. Bottom-region long sentences are
often classified as subtitle-like and suppressed from semantic summaries.
Do not treat OCR text as exact transcription unless confidence and frame quality are strong. Abstain from naming features, screens, or workflows if the supporting lines are low-confidence, suppressed, or only present in rejected frames.
The input must point to an existing storyboard manifest and image files.
Tesseract/OCR setup failures, missing frames, unreadable images, or an overly
high minConfidence can produce sparse output. Sparse output is a signal to
inspect the frame images, not a signal that the UI contains no text.
Repo-side runner:
node --import tsx scripts/harness/storyboard-ocr.ts
Installed-pack runner:
node ./.video-evaluator/agent/run-tool.mjs storyboard-ocr