On this page
  1. Install
  2. What it can do
  3. Commands
  4. Questions

Install

Before you start: Node 22 or newer, Chrome or Edge (palate takes screenshots with it) and your invite code (only to build; the checks are free).

  1. Step 1: Add palate to your agent

    Claude Code

    In a terminal:

    Terminal
    claude mcp add palate -- npx @palate/cli mcp

    Codex and Cursor read a config file instead:

    Codex

    Add to ~/.codex/config.toml:

    ~/.codex/config.toml
    [mcp_servers.palate]command = "npx"args = ["@palate/cli", "mcp"]

    Cursor

    Add to .cursor/mcp.json in your project, or ~/.cursor/mcp.json for every project:

    .cursor/mcp.json
    {  "mcpServers": {    "palate": { "command": "npx", "args": ["@palate/cli", "mcp"] }  }}

    Any other agent that runs MCP servers: the command is npx @palate/cli mcp.

  2. Step 2: Activate it

    Once per machine, with the code from your invite. A code works on up to three machines.

    Terminal
    npx @palate/cli activate palate-XXXX-XXXX-XXXX

    To check it on this machine: npx @palate/cli activate --status

  3. Step 3: Try it

    Ask your agent, in your own words:

    Claude Code

    make the pricing section

    palate_plan

    Build: one section, and you didn't ask to explore.One build from your references and rules, then the finishing pass,then you point at what's wrong. No variants (~137k tokens estimated).

    What happens next: palate sizes the work first (a fix, a build, or research), says what it will cost in tokens, then guides your agent through it.

Or let your agent do it

Paste this into your agent’s chat, with your own invite code:

Set up palate for me: read https://palate.sh/docs.md and follow the install steps for this agent. My invite code is palate-XXXX-XXXX-XXXX.

What it can do

Ask in your own words. Each of these is what to type and what happens next.

Not done until it's checked

When your agent says it's finished, palate checks first. Free, no code needed.

Type: add a testimonials row

  1. Every changed screen looked at after its last edit; the copy, credits and gates checked
  2. Sent back with what's still open, three rounds at most, then it tells you
  3. In Claude Code, Codex, Cursor, Gemini CLI and Copilot, or a git check for any agent

Command: palate guard --install

Fix or tweak

A small change, made in place and checked once. No loop.

Type: the headings are three different sizes

  1. Changed where it is, in your code
  2. One screenshot of the result, then palate check
  3. About 12k tokens

Command: palate plan → direct

Build a page

One build, planned before code and finished to the last state.

Type: build the pricing page

  1. A blueprint first: the system, a look of its own, one archetype per section
  2. The hero and every section built, then every state and detail
  3. Judged by agents that didn't build it; about 137k tokens

Command: palate build

Something new

For a new product or a new kind of screen: research before the build.

Type: design a homepage for a bakery that takes weekend pre‑orders

  1. Competitors and well-crafted pages looked at live
  2. Three art directions; palate picks from your taste, or you do
  3. Then one build; about 174k tokens

Command: palate researchpalate direction

Match my design

Your screenshot, page or Figma frame is the spec, built faithfully.

Type: build this from my screenshot

  1. Every value tagged measured, guessed or swapped
  2. Your design and the build checked side by side
  3. A verdict: ready, ready with substitutions, or not yet

Command: palate match

Point at what's wrong

Click anything in your running app and say what should change.

Type: let me point at what's off

  1. palate opens your app with its overlay
  2. Your agent fixes each pin while you keep going
  3. A before and after for every fix, at both widths

Command: palate point

Options, when you ask

Several directions as sketches, only when you want to explore.

Type: show me three takes on the hero

  1. Sketches built in parallel
  2. You pick blind, side by side, in your browser
  3. The winner is built out in full

Command: palate variantspalate pick

Critique

A designer's review of a page, section by section.

Type: review this page like a designer

  1. Findings on a board in your browser
  2. Agree or disagree with each one
  3. Agreed ones get fixed; the rest teach palate

Command: palate critiquepalate fix

Your references and rules

Tell it what you like and what you never want. It keeps both.

Type: never use cream backgrounds

  1. Kept as a rule, at the top of every brief
  2. Checked on every screenshot and by every judge
  3. Every pick, pin and correction can become a rule you keep, by type of screen

Command: palate rule addpalate refs addpalate taste

Commands

Everything the MCP server does is also a command. Your agent calls the MCP tool; you can run the command yourself.

palate --help
37 commands
Every palate command, grouped by what it is for, with the MCP tool your agent calls for it
CommandWhat it doesMCP toolCopy
Check, free6
palate guard --install [--agent …]Your agent can't call UI work done until palate has looked at it. A stop hook for Claude Code, Codex, Cursor, Gemini CLI and Copilot; --agent git for any agent.MCPpalate_guard
palate guardWhat's still open: screens not looked at since their last edit, copy, credits, gates.MCPpalate_guard
palate copy [files…]Copy that reads as generated (hype, filler, placeholder text, proof with no source), with what to write instead.MCPpalate_copy
palate waive <id> --reason "…"A finding that's wrong for your project, with the reason. Every report lists it.MCPpalate_waive
palate imagery "<query>" [--sources commons]Photos you may use: CC0 museum images, or Commons photos under CC0, public domain or CC BY, credited.MCPpalate_imagery
palate icons <brand…>Real brand marks as local SVGs in their own colour, for a works-with row.MCPpalate_icons
Start5
palate activate <code>Activate palate on this machine with your invite code, once.MCPpalate_activate
palate activate --statusWhether palate is activated here.CLI only
palate plan "<request>" [--route /x]What a request deserves (a fix, a build, research, options), why, and its token cost.MCPpalate_plan
palate contextWhat the project already uses: stack, tokens, fonts, components, screens.MCPpalate_context
palate versionThe installed version.CLI only
Build7
palate build "<job>" --route /xOne build of a real route from your rules, references and taste.MCPpalate_build
palate blueprint --file <plan.json>Check a new page's plan before any code: system, look, one archetype per section.MCPpalate_blueprint
palate research "<job>"Competitors and well-crafted pages, looked at live, for a new screen.MCPpalate_research
palate direction "<job>"Three art directions, or three ideas for a problem, before a build.MCPpalate_direction
palate match <image | url | figma link>Build your own design faithfully, then check it side by side.MCPpalate_match
palate variants "<job>" --name <n>Several directions as sketches, when you ask for options.MCPpalate_variants
palate pick --name <n>Pick the best variant blind, side by side, in your browser.MCPpalate_pick_start
Finish9
palate finish <route>Screenshots at both widths, every state, the small details and motion, in one pass.MCPpalate_finish
palate gate <route>The quality floor, judged by fresh agents that didn't build the page.MCPpalate_gate
palate shoot <route>Screenshots and measured problems: contrast, overflow, small text, repeats.MCPpalate_shoot
palate states <route>Empty, loading, error, long text, offline and keyboard states, rendered.MCPpalate_states
palate details <route>The small parts: avatars, inputs, icons, badges, hover and focus.MCPpalate_details
palate animate <route> --checkMotion that serves the content, checked frame by frame.MCPpalate_animate
palate critique <route>A designer's review, section by section.MCPpalate_critique
palate fix <route>A brief from the findings you agreed with, most severe first.MCPpalate_fix
palate checkGenerated-UI tells and token drift in your changed code.MCPpalate_check
Point2
palate point [route]Open your running app with palate's overlay; click what's wrong and say why.MCPpalate_point_start
palate point [route] --autoThe same, and palate runs your agent headless on each pin.CLI only
Your taste7
palate refs add <image | url> --like "…"Save a reference with what you like and dislike about it.MCPpalate_refs_add
palate rule add "<rule>"A rule in your words, active at once and checked on every build.MCPpalate_rule
palate rule reviewKeep or discard the rules palate proposed from your picks and fixes.MCPpalate_rule
palate taste [learn]Your taste profile, by type of screen; learn proposes new rules.MCPpalate_taste
palate fail <route> "<why>"Mark a screen as bad and why; palate proposes rules from it.MCPpalate_fail
palate forget <project>Drop a throwaway project from palate's history; taste is untouched.CLI only
palate statsTime and tokens per mode, from your own runs.MCPpalate_stats
Run as a server1
palate mcpRun palate as an MCP server (stdio) for Claude Code, Codex, Cursor and other MCP clients.CLI only

In Claude Code you can also use the slash prompts

Questions

What's free, and what needs an invite?

Checking and finishing screens you already have is free with no code: the guard, the copy check, screenshots and every state, the quality gate, photos and brand icons. Building from your taste, research and directions, options, pointing at what's wrong, critique and the rules palate learns need an invite. What palate learns while you use the free part is there when you activate.

npx @palate/cli guard --install

Does my code or design leave my machine?

No. Screenshots, references, picks and your taste profile stay on your machine. The only network call is activation and an occasional licence check, which send your invite code and an anonymous machine id.

~/.palate

Which agents does it work with?

Any agent that runs MCP servers: Claude Code, Codex and Cursor are set up above. The guard hooks into Claude Code, Codex, Cursor, Gemini CLI and Copilot; for any other agent, a git pre-commit check does the same job. Everything the server does is also a command, so the CLI works on its own.

npx @palate/cli mcp

How many tokens does it use?

palate tells you before it starts. A fix is about 12k tokens, a page build about 137k, research and a build about 174k. Options are opt-in because they cost more.

palate stats

Does it work with my stack and components?

Yes. It reads your project first (its stack, tokens, fonts and components) and builds in your codebase with them. It starts your own dev server: Next.js, Vite, Astro, Nuxt, Remix and React Router, Angular, SvelteKit and Create React App.

palate context

How many machines can I use it on?

Up to three per invite code. Activate once on each; palate re-checks the code now and then and keeps working offline for a while if it can't reach us.

npx @palate/cli activate --status

Something broke, or I have an idea?

Reply to your invite email, or write to Garvit through garvit.me. Bugs with the command you ran and what you saw get fixed fastest.

garvit.me