Prompts / token_audit
Design token and colour audit Free
Checks colour, type and spacing tokens for naming, duplication, scale discipline and contrast, using the exact values from the file.
Run it
- Claude Code:
/mcp__designers__token_audit - Claude Desktop and claude.ai: plus menu, Add from Designers, Design token and colour audit
- Any app: Use the token_audit prompt from the Designers server on figma link to a file with variables, or the tokens pasted as css variables or json.
Arguments
| Name | Meaning |
|---|---|
source * | Figma link to a file with variables, or the tokens pasted as CSS variables or JSON |
* required
What your assistant is told to do
The full instructions, so you can see exactly what you are asking for:
Audit these design tokens.
Source: <source>
Method:
1. If a Figma link: If the input is a Figma link: parse it (figma.com/design/:fileKey/..., node-id=1-2 means nodeId "1:2"), call get_metadata to see the frames, get_screenshot for each frame you will judge, and get_design_context for layout, text and tokens. Read the picture before the code. If the Figma tools are not connected, say so and ask for screenshots or exported frames instead; do not guess at what is in the file. Call get_variable_defs for the collection and modes. If tokens are pasted, parse them.
2. Naming: are tokens semantic (color.text.primary, color.surface.raised) over literal (blue-500) at the component layer, with a primitive layer beneath? List tokens named by value or by location.
3. Duplication: find near-duplicate colours (within about 3 percent), sizes that differ by 1px, spacing off the base unit. Propose merges.
4. Scale: spacing on a 4 or 8 base, type on a ratio, radius from a short set.
5. Contrast: pair every text colour with every background colour it can sit on. Call check_contrast with pairs and the palette. Use check_contrast_tokens to run the whole matrix in one call (free up to about 30 colours, Pro for 200). Report failing pairs (4.5:1 text, 3:1 large text and components) with the nearest passing colour, and colour pairs that merge under colour blindness.
6. Modes: light and dark as value sets over the same semantic names, with contrast checked in each.
Output: a short scorecard (naming, coverage, scale, contrast, modes) with evidence, a table of proposed merges and renames, the failing pairs with fixes, and the ten highest-impact changes.