agentleFS
45ck45ck/video-evaluator/skills/storyboard-ocr/SKILL.md

storyboard-ocr

Skill2 starsChanged 7 months agoMITView on GitHub

OCR extracted storyboard frames, write a text manifest, and surface evidence-backed UI text from a local video.

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 1 is distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Use this after storyboard-extract when an agent needs machine-readable UI text from extracted frames instead of only images.

    line 8
  2. Very specific:

    Run after storyboard-extract.

    line 48
  3. Very specific:

    Use semanticLines for product/UI claims.

    line 54
  4. Distinctive:

    Do not treat OCR text as exact transcription unless confidence and frame quality are strong.

    line 61

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Storyboard OCR

Use this after storyboard-extract when an agent needs machine-readable UI text from extracted frames instead of only images.

This skill owns:

  • OCR over extracted storyboard frames
  • per-frame text with confidence filtering
  • per-line boxes and coarse regions when block/layout OCR is available
  • storyboard.ocr.json
  • an aggregated unique-text summary for later review prompts

Inputs

Provide 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.

Outputs

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.

Sequencing Guidance

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.

Interpretation Notes

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.

Abstention Rules

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.

Failure Modes

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

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.