Agent skill

Workflow

by ambient-code in ambient-code/platform

Perform any development work on the ambient-ui component. An agent skill from ambient-code/platform.

MITAuto-check passedFrontend & Design

Install Workflow

skills CLI
$ npx skills add ambient-code/platform --skill workflow -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install ambient-code/platform workflow --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/ambient-code/platform.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/ambient-ui/workflow .claude/skills/workflow && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
workflow
GitHub stars
131
Token cost
~2.9k tokens
SKILL.md length
1,422 words
Files
2
Skills in repo
18
Repo updated
First seen
Licence
MIT

At a glance

Perform any development work on the ambient-ui component. An agent skill from ambient-code/platform.

  • Works in 5 steps: Align with Specs → Perform Test Driven Development → Perform Critical Review → …
  • Frontend & Design work in your project
  • SKILL.md covers User Input and Workflow
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Workflow is an agent skill from ambient-code/platform. Perform any development work on the ambient-ui component. Full workflow instructions.

Its SKILL.md is about 2.9k tokens, which your agent loads only when the skill is triggered. The skill folder holds 2 other files (for example `evals/evals.json`).

It sits in Frontend & Design. The repository describes itself as: Vision/Mission https://ambient-code.ai : Virtual team management and collaboration platform. User guides: https://ambient-code.github.io/platform/. The licence is MIT.

When your agent uses it

  • Frontend & Design work in your project

Example prompts

  • “/workflow”

Workflow steps

5 steps, taken from the step headings in SKILL.md.

  1. Align with Specs
  2. Perform Test Driven Development
  3. Perform Critical Review
  4. Address Review Comments
  5. Request User Feedback

What it can do on your machine

Read from SKILL.md and the folder at commit e15afd3. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Workflow loads about 2.9k tokens when it runs. Until then it costs about 24 tokens; SKILL.md has 1,422 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~24
When it runs · the whole SKILL.md, loaded when a task matches
~2.9k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from ambient-code/platform at commit e15afd3, republished under its MIT licence (© ambient-code). 1,422 words, ~2,922 tokens.

Download SKILL.mdSave it as .claude/skills/workflow/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
workflow
description
Perform any development work on the ambient-ui component. Full workflow instructions.

Ambient UI Development workflow

Strict workflow execution to be used when performing any ambient-ui component work.

User Input

text
$ARGUMENTS

Workflow

1. Align with Specs

All work must be aligned with specs. If the user or instructions are asking for work to be done without a verified matching section in a spec, STOP.

Before proceeding with implementation, update the relevant spec to match the request.

2. Perform Test Driven Development

Determine the size of the work. If it is large, break it down into smaller atomic units of work and execute this workflow for each one. You can use subagents to handle this gracefully, preserving your context.

Define tests that are relevant and that, when they pass, provide a genuine trust that the system is properly implemented in a way that aligns with the spec and the user's intent.

Then, proceed with methodical implementation, breaking work into atomic units of work for subagents when possible. Bias towards using subagents for implementation tasks, but ensure they read this workflow skill to understand the development standards.

Test and verify implementation before proceeding to the next step.

Standards

These standards are to be used during all ambient-ui component development work.

Fakes over Mocks

When testing, fakes are to be preferred over mocks. Tests that don't test reality are not tests at all.

Accessibility
  • For Red Hat assets, small text (17pt or smaller) needs to have a 4.5:1 contrast ratio at minimum. Large text (18pt or larger) and informative graphics like icons need to have a 3:1 contrast ratio at minimum.

  • Saturated hues that are similar in intensity can vibrate when used together, creating fuzziness around the edges like a glowing blur. These colors can be difficult for anyone to look at and distinguish, and even painful for those with vision differences. An example of high-contrast colors that do not vibrate.

    Do this: Combine bright colors with less saturated and neutral colors. Image showing misuse: An example of low-contrast colors that vibrate.

    Not this: Don’t use colors with similar intensity in the same area.

Technology
  • React/Next.js
  • shadcn/ui components
  • Tabbed views MUST persist the active tab in the URL via ?tab= query parameter using controlled <Tabs value={activeTab}> with useState + useEffect + replaceState. Ensures tabs survive reload and support deep linking.
Colors

Core Palette

Color NameHEX CodeRGBCMYKPantone / Notes
red-05#fef0f0———
red-10#fce3e3———
red-20#fbc5c5———
red-30#f9a8a8———
red-40#f56e6e———
red-50#ee0000238 0 00 98 85 0Pantone 1788C (Red Hat red)
red-60#a60000———
red-70#5f0000———
red-80#3f0000———
white#ffffff255 255 2550 0 0 0Pantone White
gray-10#f2f2f2———
gray-20#e0e0e0———
gray-30#c7c7c7———
gray-40#a3a3a3———
gray-45#8c8c8c———
gray-50#707070———
gray-60#4d4d4d———
gray-70#383838———
gray-80#292929———
gray-90#1f1f1f———
gray-95#151515——ux black
black#0000000 0 060 40 40 100Pantone Black C

Secondary Palette

Color NameHEX CodeRGBCMYKPantone / Notes
orange-10#ffe8cc———
orange-20#fccb8f———
orange-30#f8ae54———
orange-40#f5921b245 146 270 50 100 0Pantone 144C
orange-50#ca6c0f———
orange-60#9e4a06———
orange-70#732e00———
orange-80#4d1f00———
yellow-10#fff4cc———
yellow-20#ffe072———
yellow-30#ffcc17248 204 230 15 100 0Pantone 108C
yellow-40#dca614———
yellow-50#b98412———
yellow-60#96640f———
yellow-70#73480b———
yellow-80#54330b———
teal-10#daf2f2———
teal-20#b9e5e5———
teal-30#9ad8d8———
teal-40#63bdbd———
teal-50#37a3a355 163 16380 10 30 10Pantone 2234C
teal-60#147878———
teal-70#004d4d———
teal-80#003333———
purple-10#ece6ff———
purple-20#d0c5f4———
purple-30#b6a6e9———
purple-40#876fd4———
purple-50#5e40be94 64 19085 80 0 0Pantone 2097C
purple-60#3d2785———
purple-70#21134d———
purple-80#1b0d33———

Note: The "Auxiliary colors" section in your text was empty, so it has been omitted here.


Information Palette

⚠️ Usage Note: These colors are utilitarian. They should only be used for their intended purposes, not for decorative visuals.

Color NameHEX CodeRGBCMYKPantone / Notes
success-green-10#e9f7df———
success-green-20#d1f1bb———
success-green-30#afdc8f———
success-green-40#87bb62———
success-green-50#63993d99 153 6170 0 100 10Pantone 7737C
success-green-60#3d7317———
success-green-70#204d00———
success-green-80#183301———
danger-orange-10#ffe3d9———
danger-orange-20#fbbea8———
danger-orange-30#f89b78———
danger-orange-40#f4784a———
danger-orange-50#f0561d240 86 290 83 100 0Pantone 165C
danger-orange-60#b1380b———
danger-orange-70#731f00———
danger-orange-80#4c1405———
interaction-blue-10#e0f0ff———
interaction-blue-20#b9dafc———
interaction-blue-30#92c5f9———
interaction-blue-40#4394e5———
interaction-blue-50#0066cc0 102 20485 55 0 5Pantone 2387C
interaction-blue-60#004d99———
interaction-blue-70#003366———
interaction-blue-80#032142———

Color Association More info red Red Hat Red is our brand color. Do not use red to represent negative things. danger-orange Error, decrease, or failure Something negative has occurred, like a destructive error or a decrease in value. orange Caution A non-destructive action or error has occurred. yellow Warning Take action now to avoid a destructive action or error. success-green Success, increase Something positive has occurred, like a successful action or increase in value. teal General or neutral A button or information has no severity. interaction-blue Link or interaction Clicking the object or text leads to a hyperlink or state change. purple Info or note tip Helpful information is available. gray Null A button or information is unavailable or unimportant.

Show full SKILL.md (612 more words)Show less
Typefaces

Red Hat Text takes all of the personality from Display and optimizes it for more demanding applications. As the name implies, it’s easier to read in paragraphs or when text is used at small sizes—like in a whitepaper or tooltip.

To increase readability, Text has more height difference between the upper- and lowercase letters, more space between narrow characters, and more variation in letterform stroke weights.

Red Hat Mono

Red Hat Mono was created to distinguish code from natural-language text. Mono stands for monospaced, meaning each letter takes up the same amount of horizontal space. This creates neat columns of text and makes scanning code easier.

Mono should only be used when demonstrating code snippets in communications and interfaces, or as a stylistic approach for a more technical audience like the Red Hat Developer Program or the Code Comments podcast.

General Principles
  • Use consistent design tokens
  • Use iconography to improve the user experience and ability to scan the page
  • Everything must be accessible
  • Use reusable components
  • Unless specified by the user, NEVER MAKE CONSIDERATIONS BASED ON A NEED FOR BACKWARDS COMPATIBILITY.
  • Use atomic, conventional commits.
3. Perform Critical Review

Launch the following agents in parallel to review and critique your work:

  1. Security review agent:
  • Expert in OWASP vulnerability classes
  • Thoroughly vets for supply chain vulnerabilities
  1. Spec alignment review agent:
  • Thoroughly review the relevant spec & the implementation. Flags any gaps, no matter how trivial. Spec is the source of truth.
  1. Architecture review agent:
  • Identifies architectural inconsistencies, deviations from standards (domain driven design, domain oriented observability, using ports/adapters, etc.)
  • Identifies duplicate code, magic strings/numbers, etc.
  1. Code quality agent:
  • Identify lint & formatting issues
  • Identify dead code
  1. Documentation sync agent:
  • Ensure relevant README & documentation reflects the changes in this unit of work
  • Do not be overly verbose. Follow any project documentation guidelines, aim for conciseness. Do not over document.
  1. UX review agent:
  • Embodies the persona of Steve Krug and the book "Don't Make Me Think"
  • Identifies anything that might make the user think, is poorly organized, doesn't follow principles, etc.
  • Uses playwright to see and interact with the ui to verify & discover all findings.
  1. User target demographic review agent:
  • Embodies the persona of an engineer that is using the platform to:
    • Observe units of work, their status, and the overall status of the system (is human intervention necessary etc.)
    • Define and maintain credentials, binding access to projects and specific agents.
  • Exercises the application from end to end, following real user patterns to discover system breaking bugs.
4. Address Review Comments

Use subagent(s) to address relevant review comments. Although you are to use your judgement about whether a review comment is relevant, you are also to be picky about your work. Take pride in doing the correct thing, even if it's not the easiest.

After addressing review comments, loop back to step 3.

If step 3 provides no actionable comments, proceed to the next step. Do not worry if you spend many cycles in this stage. Correctness is worth the effort.

5. Request User Feedback

Get the UI in a running state so that the user can provide feedback. There are two ways for you to do this:

Running inside ambient

If you are running within the ambient platform, you will have instructions for building images, and deploying to an OpenShift cluster. You should do this, create a route, and add the route to the relevant annotation (see your operating instructions.)

Running locally

If you are running locally, simply follow the local development setup, and provide the user with the URL to open in their browser to provide feedback. Let them know what to test and where to look.

© ambient-code, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file in skills/ambient-ui/workflow of ambient-code/platform.

  • SKILL.md
  • evals/evals.json

Open the folder on GitHubat commit e15afd3

Compare with similar skills

Workflow next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Workflow compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Workflow this skillambient-code/platform131—~2.9kAutomated safety check: PassMIT
Web Artifacts Builderanthropics/skills180k40 repos~769Automated safety check: PassApache-2.0
React Doctormakeplane/plane60k12 repos~657Automated safety check: PassAGPL-3.0
React Composition Patternsvercel-labs/openreview1.7k58 repos~721Automated safety check: PassMIT
Impeccablebestofjs/bestofjs3.1k27 repos~2.6kAutomated safety check: PassMIT
Figma Design System Builderwarpdotdev/warp65k2 repos~4.4kAutomated safety check: PassAGPL-3.0

Similar skills

  • Web Artifacts Builder

    anthropics/skills

    Official

    Builds multi-component claude.ai HTML artifacts as a small React, TypeScript and Tailwind project, then bundles it into one shareable HTML file.

    180k GitHub starsUsed in 40 repos~769 tokens
    Frontend & DesignAuto-check passed
  • React Doctor

    makeplane/plane

    Scans React code for lint, accessibility, bundle size and architecture issues, reports a health score and checks that changes do not lower it.

    60k GitHub starsUsed in 12 repos~657 tokens
    Frontend & DesignAuto-check passed
  • React Composition Patterns

    vercel-labs/openreview

    Official

    Rules for structuring React components with composition instead of boolean props, covering compound components, lifted state, variants and React 19 changes.

    1.7k GitHub starsUsed in 58 repos~721 tokens
    Frontend & DesignAuto-check passed
  • Impeccable

    bestofjs/bestofjs

    A skill your agent uses when the user wants to design, redesign, shape, critique, audit, polish, clarify, distill, harden, optimize, adapt, animate, colorize, extract, or otherwise improve a…

    3.1k GitHub starsUsed in 27 repos~2.6k tokens
    Frontend & DesignAuto-check passed
  • Builds or updates a design system in Figma from a codebase in ordered phases: discovery, variables and tokens, components, theming and documentation, with checkpoints.

    65k GitHub starsUsed in 2 repos~4.4k tokens
    Frontend & DesignAuto-check passed
  • Tailwindcss Development

    anonaddy/anonaddy

    Always invoke when the user's message includes 'tailwind' in any form.

    4.9k GitHub starsUsed in 10 repos~865 tokens
    Frontend & DesignAuto-check passed

More from ambient-code/platform

All 18 skills in this repo
  • Align

    ambient-code/platform

    Run a convention alignment check across the codebase to measure adherence to documented standards.

    131 GitHub stars~754 tokensUpdated yesterday
    Auto-check passed
  • Amber Review

    ambient-code/platform

    Perform a comprehensive code review using repository-specific standards.

    131 GitHub stars~1.4k tokensUpdated yesterday
    Auto-check passed
  • Cypress Demo

    ambient-code/platform

    Create a Cypress-based video demo for a feature branch with cursor, click effects, and captions.

    131 GitHub stars~879 tokensUpdated yesterday
    Auto-check: notes
  • Deploy

    ambient-code/platform

    Deploy, update manifests, and troubleshoot the ambient-ui component.

    131 GitHub stars~2.6k tokensUpdated yesterday
    Auto-check passed
  • Discover

    ambient-code/platform

    Discover domain-specific skills and context before starting any coding or spec work.

    131 GitHub stars~747 tokensUpdated yesterday
    Auto-check passed
  • Frontend Development

    ambient-code/platform

    Required context for all frontend work. An agent skill from ambient-code/platform.

    131 GitHub stars~525 tokensUpdated yesterday
    Auto-check passed

Questions about Workflow

What does Workflow do?

Perform any development work on the ambient-ui component. An agent skill from ambient-code/platform. Workflow is an agent skill from ambient-code/platform. Perform any development work on the ambient-ui component.

When should I use Workflow?

Workflow fits situations like: frontend & Design work in your project.

How do I install Workflow in Claude Code?

Run `npx skills add ambient-code/platform --skill workflow -a claude-code`. Or copy the skill folder (skills/ambient-ui/workflow in ambient-code/platform) into .claude/skills/workflow in your project. Claude Code loads it when a task matches its description.

How do I install Workflow in Codex?

Run `npx skills add ambient-code/platform --skill workflow -a codex`. Or copy the skill folder (skills/ambient-ui/workflow in ambient-code/platform) into .agents/skills/workflow in your project. Codex loads it when a task matches its description.

Can I use Workflow in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add ambient-code/platform --skill workflow -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/workflow, .gemini/skills/workflow, .github/skills/workflow and .opencode/skills/workflow in your project.

What does Workflow need to run?

SKILL.md names no scripts, command-line tools or credentials: Workflow is instructions for the agent only.

Does Workflow access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Workflow safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Workflow use?

Workflow is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Workflow use?

About 2.9k tokens (SKILL.md is roughly 12k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Workflow?

Skills that share tags, products or a category with Workflow: Web Artifacts Builder (anthropics/skills, 180k stars), React Doctor (makeplane/plane, 60k stars), React Composition Patterns (vercel-labs/openreview, 1.7k stars) and Impeccable (bestofjs/bestofjs, 3.1k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Workflow?

ambient-code (a GitHub organization) maintains it in ambient-code/platform, which has 131 GitHub stars. The repository holds 18 skills in this directory. The repository was last updated on October 6, 2026.

Source: ambient-code/platform on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.