nlds-update

name	nlds-update
description	Update an existing NL Design System token set — re-extracts branding from website/huisstijlhandboek and applies changes

Update an existing NL Design System token set for: $ARGUMENTS

Instructions

Gather required information

If $ARGUMENTS is empty or does not contain an organisation name, ask:

1. "What is the name of the organisation to update?" (required — do not proceed until answered)

Once you have the name, verify the token set exists at municipalities/<slug>-design-tokens/. If it does not exist, tell the user and suggest using /nlds:new instead.

Then ask:

2. "Should I re-extract the design from the organisation's website, a huisstijlhandboek (style guide), or both?"

If the user says website or both, look up the website URL from the existing package.json (the "website" field) and ask:

3. "Is this still the correct website: <url>? (or provide the right one)"

If the user says huisstijlhandboek, ask:

4. "What is the URL of the huisstijlhandboek?"

Derive identifiers

From the existing token set, read src/config.json to get:

• slug: the prefix field
• fullName: the fullName field
• prefix: same as slug
• packageName: from package.json name field
• folderName: municipalities/<slug>-design-tokens

Allowed references

Use only these sources:

1. The organisation's website
2. The organisation's huisstijlhandboek (style guide)
3. The conduction-design-tokens/ baseline files (for file structure and default values)
4. The organisation's own existing token set (the one being updated)

Do NOT look at other themes in municipalities/ for reference (e.g. to see how they chose values or mapped sizes). Other themes contain organisation-specific choices and mistakes that silently leak into this theme — for example copying another theme's decision to map the body font-size to font-size.md instead of the website's actual <p> font-size.

Execute

Read the shared reference at conduction-theme/.claude/commands/nlds-reference.md and follow it, skipping Part B (file creation):

1. Extract (Part A) — follow the extraction procedure (website and/or huisstijlhandboek as the user requested)
2. Skip Part B — files already exist, do not recreate them
3. Update component tokens (Part C) — read existing token files, compare with new extraction data, update only values that changed
4. Update brand tokens — if the extraction reveals new/changed colors or fonts:
   • Update color.tokens.json with any new palette entries
   • Update typography.tokens.json if the font changed
   • Update font-size.tokens.json if sizes changed
   • Download any new font files to src/font/ and update font.scss
5. Build and test (Part D) — follow the build procedure, fix any errors
6. Summary (Part E) — output the summary, highlighting what changed vs the previous version