scribe-import

---

name: scribe-import description: Convert Scribe markdown export into a tutorial page under /learn/getting-started. Use when the user pastes Scribe-exported markdown and wants it turned into a site page. disable-model-invocation: true argument-hint: "" "" allowed-tools: Bash, Read, Write, Edit, Glob, Grep

Scribe Import

Convert a Scribe markdown export into a properly formatted tutorial page on this site.

Input

• $0 = Card title (e.g., "What Are Organisms")
• $1 = Card subtitle (e.g., "Species and taxa served by BRC Analytics")
• The user will paste raw Scribe markdown in the conversation

If the user hasn't pasted the Scribe markdown yet, ask them to paste it.

Steps

1. Derive slug

Convert the title to a URL slug: lowercase, spaces to hyphens, strip punctuation. Example: "What Are Assemblies" → what-are-assemblies

2. Download images

• Extract all image URLs from the Scribe markdown (typically colony-recorder.s3.amazonaws.com URLs)
• Create directory: public/learn/getting-started/<slug>/
• Download each image with a short descriptive filename (e.g., filter-panel.jpeg, click-organisms.jpeg)
• Use curl -sL -o <path> <url> for each download
• Verify all downloads succeeded by checking file sizes

3. Create MDX page

Create app/docs/learn/getting-started/<slug>.mdx with this structure:

Frontmatter:

---
breadcrumbs:
  - path: "/learn"
    text: "Learn"
  - path: "/learn/getting-started"
    text: "Getting Started"
  - path: ""
    text: "<Title>"
contentType: "ARTICLE"
description: ""
enableOutline: true
title: "<Title>"
---

Body — rewrite each Scribe step as:

## Short descriptive heading

One or two sentences describing what the user sees or should do. Fix any grammar issues from the original.

<Figure src="/learn/getting-started/<slug>/<image-filename>" alt="Descriptive alt text" />

Rules:

• Each step gets a short ## heading (2-4 words) followed by descriptive text
• Use <Figure> component for all images (not markdown ![]() syntax)
• Image src paths start with /learn/ (not /public/learn/)
• Remove ALL Scribe attribution links and "Made with Scribe" references
• Fix grammar, spelling, and style issues in the original text
• Merge consecutive steps that describe the same action into one section
• Use italics for species names

Reference these files for style guidance:

• app/docs/learn/getting-started/what-are-organisms.mdx
• app/docs/learn/getting-started/what-are-assemblies.mdx

4. Add card to getting-started page

Read app/docs/learn/getting-started.mdx and append a new entry to the cards array in frontmatter:

  - href: "/learn/getting-started/<slug>"
    secondaryText: "<subtitle from $1>"
    title: "<Title from $0>"

5. Verify

• Confirm the MDX file exists and has valid frontmatter
• Confirm all images exist in public/learn/getting-started/<slug>/
• Confirm the card was added to getting-started.mdx
• Tell the user to check the page at http://localhost:3001/learn/getting-started/<slug>