---
title: "Is there a Claude Code skill that turns a script into a storyboard?"
description: "Yes, Wireflow's storyboard skill for Claude Code turns a script into six 16:9 frames with one character and one look, for 60 credits per run."
updated: 2026-11-07
---

Yes, Wireflow's storyboard skill for Claude Code turns a script, an ad brief or a pitch idea into six 16:9 frames with one character and one look. It breaks the story into six shots, writes one character note and one visual style, quotes the credits and waits for your yes, then renders all six frames in one call with Seedream 5 Lite and saves them to `./wireflow-outputs` with a shot list.

It is made for an agency, a brand team or a solo creator who needs to show a short video before anything is shot or generated. The planning happens in Claude, and the frames render as a published Wireflow app over the Wireflow MCP connector, signed in as you. Without the connector, the skill still writes the shot list and frame prompts. For finished static ads rather than a plan, use the [ad creative skill](/docs/skills/ai-ad-creative-generator).

Its instructions and storyboard planner are in [the skill's folder on GitHub](https://github.com/wireflowINC/skills/tree/main/skills/ai-storyboard-generator).

## How do I install the storyboard skill in Claude Code?

Install the Wireflow plugin, which adds every Wireflow skill and the Wireflow connector together:

```text
/plugin marketplace add wireflowINC/skills
/plugin install wireflow-skills@wireflow
```

Claude Code shows the plugin's details first, so you can review it and choose its scope. Then run `/mcp`, choose `wireflow` and sign in with your Wireflow account. Claude picks skills up from plain requests, and you can also call a skill by name as a slash command.

To add only this skill, use the skills CLI, or copy its folder from under `skills/` in the repo into `~/.claude/skills/` (all projects) or `.claude/skills/` (one project). Then connect Wireflow yourself and run `/mcp` to sign in. The connector uses OAuth, so there is no API key to copy.

```bash
npx skills add wireflowINC/skills --skill ai-storyboard-generator
claude mcp add --transport http wireflow https://www.wireflow.ai/api/mcp
```

To check that it worked, ask Claude for a storyboard. The skill first looks for the Wireflow tools `run_app`, `get_execution`, `get_credit_balance` and `whoami`; when they are there, it goes on to ask for your script, hero and look. When they are missing, it says so, shows how to connect, and still gives you the free part: the shot list filled in for your idea, with the style and character blocks, ready to sketch by hand or paste into any image tool.

## What do I tell Claude to plan a storyboard?

Tell Claude the script, voiceover, brief or idea, how long the finished video is (15, 30 or 60 seconds), who the hero is and what they wear, and the look you want. A sketch suits an internal pitch, while painted or photoreal suits a client deck or an AI video brief.

```text
Storyboard this 30 second video ad for a client deck. Script: a hiker
reaches a cold summit at dawn, pours tea from a steel flask and watches
the sun come up. Hero: a woman in her 40s with a grey braid, a red down
jacket and a black wool hat. Look: painted, soft dawn light. It runs as
a 16:9 web ad.
```

The skill is meant for requests to storyboard a video ad, a commercial, a client pitch, or a TikTok or Reels concept, and for a shot list with pictures, frames to plan a shoot or panels to brief an AI video model. If you have no hero in mind yet, it offers two options that fit the audience. Mention the format the video will run in; the frames are 16:9 either way.

Inputs are text only. A real person's photo, a product shot or a logo cannot be used, so describe the product in plain words and leave brand marks for the edit. Real, identifiable people, celebrities and other brands' logos stay out of the frames.

From your request, Claude plans the six shots before anything renders:

1. Fits the story into six beats: open, problem, turn, help or product, payoff, end shot.
2. Varies the camera, with at least one wide, one close-up and one medium shot, and never two identical framings in a row.
3. Writes each panel as one or two sentences that name camera distance and angle first, then action, place and light.
4. Leaves out words, signs, screens with text and brand names, because the frames are wordless by design.
5. Shows the plan as a table (panel, shot, action, timing, voiceover line) with the style and character notes under it.

Change anything you want at this point: a frame cannot be changed after the run without running all six again.

The plan becomes eight text inputs for the Wireflow app, and the skill sends all eight every time:

| Input | What it holds |
| --- | --- |
| `style` | One paragraph for all six frames: medium (pencil sketch, ink and marker, digital painting, 3D, photoreal), palette, line and light, grain |
| `character` | The hero as they look in every frame: age, face, hair, skin, build, glasses, the full outfit with colours |
| `panel_1` | The opening shot, which sets the look for the rest: a clear wide or medium shot with the hero's face visible |
| `panel_2` to `panel_6` | One shot each: camera distance, angle, action, setting, light |

An input left out falls back to the board's saved demo, a woman in a yellow robe making coffee, and that scene would land in your storyboard.

## What do I get back from a run?

You get six landscape 16:9 JPG frames saved in order under `./wireflow-outputs/`, plus a `storyboard.md` shot list that pastes into a deck or a doc.

| What | Details |
| --- | --- |
| Frames | Six landscape 16:9 JPGs, `frame-1` to `frame-6` |
| Frame size in the skill's test | Panel 1 at 3072x1728, panels 2 to 6 at 3416x1920 |
| Files | Saved in order to `./wireflow-outputs/YYYY-MM-DD/storyboard/`, each named with a project slug and the panel number |
| Shot list | `storyboard.md` next to the images: panel number, image, shot, action and voiceover line |
| Text in the frames | None: no captions, numbers, arrows, speech bubbles, logos or borders |
| Final report | The files, the CDN links and the credits used |

Panel 1 renders first and the other five follow together; in the skill's test the run finished in a few minutes. Claude polls `get_execution` every 5 seconds until the status is no longer `PENDING`, `RUNNING` or `IN_PROGRESS`. If the run ends `FAILED`, `CANCELLED` or `TIMEOUT`, the skill stops and quotes the error word for word.

After the report, the skill offers a second look (sketch versus painted) as one more quoted run, using the frames as start images for an AI video model one shot at a time, and the board in Wireflow so you can change the size or add steps. To animate a frame, Wireflow's [video skill](/docs/skills/ai-video-generation) turns a start image into a 5 second, 16:9 clip, and an image you made with Wireflow works directly.

Treat the frames as concept art when they go to a client: they are a plan, not the final ad.

## Does the same character stay in every frame?

Yes for the hero: the app paints panel 1 first, then paints panels 2 to 6 with panel 1 as the reference, so the main character and the style carry through while each panel gets a new composition. The skill keeps the hero clearly in panel 1, face visible, so the later panels have something to match.

Other people are not locked. The app only locks the hero from panel 1, so the skill describes every other person in full (age, hair, clothes) in every panel they appear in. In the skill's test, a mechanic described once came back as a different person two panels later.

Before calling the board done, the skill looks at every frame:

- The hero matches panel 1: face, beard or hair, skin, outfit colours.
- Each frame shows the planned shot: distance, angle, action.
- Side characters match their description in every panel.
- No stray text, numbers, logos or split panels.

It names each miss per panel. Small ones, such as a prop in the wrong place, are fine for a planning board. A wrong person or a missing action needs another run of 60 credits, after your yes, with that panel rewritten.

## How many credits does one storyboard cost?

One storyboard costs 60 credits per run, 10 per frame, measured with Seedream 5 Lite on 21 Oct 2026.

| What | Credits |
| --- | --- |
| Shot plan, style and character notes, and the quote | 0 |
| One run: six 16:9 frames (Seedream 5 Lite, measured 21 Oct 2026) | 60 credits per run, 10 per frame |
| A redo, even of one frame | Another 60, only after your yes |

Before anything runs, the skill checks your balance and quotes the cost, for example "1 run x 60 credits = 60 credits for 6 frames", next to your balance. The app always renders all six frames, which is why a redo costs the full 60. If your balance is short, the skill stops and says so. It spends credits only after a yes to the exact plan and price, and afterwards reports the credits used as your balance before minus your balance after.

Rendering needs a Wireflow account with credits: see [pricing](https://www.wireflow.ai/pricing) for what each plan includes.

## Does the storyboard skill work in Cursor, Codex or Claude Desktop?

Yes: the skill is written for Claude Code and Claude, and the skills CLI also installs it in Cursor, Codex and other agents that read `SKILL.md`, but each agent must be connected to the Wireflow MCP server before the skill can render frames. Wireflow has not tested its skills in Cursor, Codex or other agents itself.

In Claude.ai or Claude Desktop, add `https://www.wireflow.ai/api/mcp` as a custom connector. Saving the frames as files needs a shell, and the skill's final report also lists the CDN links.

Connection steps: [Cursor](/docs/generate-images-and-videos-in-cursor), [Codex](/docs/generate-images-and-videos-with-codex), and every client including ChatGPT in [Use Wireflow from your agent](/docs/use-wireflow-from-your-agent). Wireflow's other skills are on the [agent skills page](/docs/skill). To run your own Wireflow workflows from code with an API key instead of an agent, see the [run API](/docs/api/run).

Last updated: 2026-11-07

---

Documentation index: fetch https://www.wireflow.ai/llms.txt for the full list of Wireflow docs.