Skip to content
Agentmart
Submit a tool

narration-to-visual-prompts

Use when the user already has script, narration, or storyboard text and needs aligned image or video prompts that are generation-ready and mapped one-to-one…

Skill by pipixia-labs

IndexedMIT★ 37
Write a review

Chalk Labs

Need help deploying narration-to-visual-prompts?

Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.

Book a call

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/pipixia-labs-narration-to-visual-prompts

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Use when the user already has script, narration, or storyboard text and needs aligned image or video prompts that are generation-ready and mapped one-to-one to the source beats.

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md and code. Checked Nov 1, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/narration-to-visual-prompts

What's inside

Skill name
narration-to-visual-prompts
Description
Use when the user already has script, narration, or storyboard text and needs aligned image or video prompts that are generation-ready and mapped one-to-one to the source beats.
License
MIT
Instructions
100 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

# Narration To Visual Prompts

Use this skill to convert spoken or written narrative beats into prompts that can directly feed image or video generation.

## Trigger Cues

- "turn this script into prompts"
- "convert each scene into image prompts"
- "write shot prompts from my narration"
- "make generation prompts from this storyboard"

## Best For

- scene-by-scene image prompts
- shot-by-shot video prompts
- turning scripts into visual prompt packs
- adapting Chinese narration into English model prompts

## When To Use

- The user already has narration, scenes, storyboard text, or a script.
- The next step is image or video generation, but the prompts are not ready yet.
- The user needs exact one-to-one mapping between narrative beats and visual prompts.

## When Not To Use

- The user only has a loose idea and no scene structure yet.
  Use `creative-brief-to-storyboard` first.
- The user only needs a style prefix or aesthetic conversion.
  Use `style-brief-to-prompt`.
- The user explicitly asks for immediate direct generation and already gave a final prompt.

## Workflow

1. Normalize the input into discrete narration units.
2. Decide the medium:
   - default to `image` for still scenes or frame-based output
   - use `video` when motion, camera movement, or transitions matter
3. Capture style constraints from the user, brand brief, or reference assets.
4. Generate one prompt per narration unit. Keep count alignment exact.
5. Write prompts in English for model compatibility, even if the source narration is in another language.
6. Add optional negative prompt or camera notes only when they materially improve execution.
7. If the user asks for final media, feed the reviewed prompts into `ImageGenerationAgent` or `VideoGenerationAgent`.

## Recommended Expert Usage

- If the user supplies a reference image, use `ImageUnderstandingAgent` with `mode="style"` or `mode="prompt"` first.
- If the user supplies a reference video, use `VideoUnderstandingExpert` with `mode="prompt"` or `mode="shot_breakdown"` first.
- Use `TextTransformExpert` with `mode="structure"` only if the source narration is messy and needs cleanup before prompting.

## Handoff Examples

For reference-style extraction before prompt writing:

```json
{"input_path":"inbox/reference.png","mode":"style"}
```

For direct image generation after the prompt pack is approved:

```json
{"prompt":"<subject + prompt_prefix_en + prompt_en>","provider":"nano_banana","aspect_ratio":"16:9"}
```

For direct video generation after the motion prompts are approved:

```json
{"prompt":"<video prompt_en>","provider":"seedance","mode":"prompt","aspect_ratio":"9:16","resolution":"720p"}
```

## Output Contract

Return a structured prompt pack like this:

```json
{
  "style_prefix_en": "...",
  "items": [
    {
      "index": 1,
      "source_narration": "...",
      "medium": "image",
      "prompt_en": "...",
      "negative_prompt_en": "...",
      "camera_or_motion_notes": "..."
    }
  ]
}
```

## Guardrails

- Keep one prompt for each narration unit. No count drift.
- Do not mix still-image language and motion-camera language carelessly.
- Do not add visual elements that contradict the narration.
- Keep the final prompts in English, but keep surrounding explanations in the user's language when replying directly.
- If the user asks for multiple visual directions, label the variants clearly instead of blending styles into one muddy prompt.
- Prefer specific visual verbs and materials over vague adjectives like "beautiful" or "nice".

From pipixia-labs/creative-claw, MIT.

Signals

Stars
37
Forks
3
Open issues
0
Last commit (this folder)
May 12, 2026 (6 months ago)
Last release
Jun 2, 2026 (5 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

Done for you

Rather not set it up yourself?

Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.

See how we build AI agents →Read the case studies

Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.

launches shipped
50+
launches shipped
media placements
500+
media placements
founders on every account
2
founders on every account

Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block

Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.

Get narration-to-visual-prompts running in your business

Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying narration-to-visual-prompts?

CallGet help
Cite this page

narration-to-visual-prompts skill. Agent Mart. https://agentmart.dev/skills/pipixia-labs-narration-to-visual-prompts (last updated 2026-10-24)