/arn-spark-visual-sketch
This skill should be used when the user says "visual sketch", "arn visual sketch", "sketch directions", "explore visuals", "visual proposals", "try different looks", "design directions", "sketch the UI", "visual exploration", "compare styles", "show me options", "what could this
$ npx -y skills add AppsVortex/arness --skill arn-spark-visual-sketch --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
- Fires itselfAuto-invocation. Claude auto-loads it when your prompt matches the work.
- You can call itInvoke it directly when you want it.
- Slash command
/arn-spark-visual-sketch
Context preview
The summary Claude sees to decide when to auto-load this skill.
This skill should be used when the user says "visual sketch", "arn visual sketch", "sketch directions", "explore visuals", "visual proposals", "try different looks", "design directions", "sketch the UI", "visual exploration", "compare styles", "show me options", "what could this
SKILL.md
arn-spark-visual-sketch.SKILL.mdname: arn-spark-visual-sketch
description: >-
This skill should be used when the user says "visual sketch", "arn visual sketch",
"sketch directions", "explore visuals", "visual proposals", "try different looks",
"design directions", "sketch the UI", "visual exploration", "compare styles",
"show me options", "what could this look like", or wants to generate multiple
visual direction proposals as real HTML/CSS running on the scaffolded project's
dev server, iteratively selecting and refining until a final visual direction
is chosen.
version: 1.0.0
Arness Spark Visual Sketch
Generate visual direction proposals as real HTML/CSS in the scaffolded project, compare them side by side in the browser, and iteratively refine until a direction is selected. This is a conversational skill that runs in normal conversation (NOT plan mode). The primary artifacts are a **visual-direction.md** document in the Vision directory and **screenshot captures** of the selected direction in the visual grounding directory.
This skill sits between `/arn-spark-scaffold` and `/arn-spark-style-explore` in the greenfield pipeline. It produces a visual-direction.md that style-explore uses as primary input for detailed design token specification. The user sees what the product could look like before committing to a full design system.
The iterative process works in rounds: 1. **Round 1:** N proposals with distinct visual directions (default 3) 2. **Round 2+:** N expansions of the selected proposal with user-guided refinements 3. **Final:** The selected direction is captured and documented
Each proposal runs on the scaffold's actual dev server using the project's real CSS framework and component library.
Prerequisites
The following artifacts inform the sketch generation. Check in order:
Determine the output directories: 1. Read the project's `CLAUDE.md` and check for a `## Arness` section 2. If found, extract the configured Vision directory path and Visual grounding directory path 3. If no `## Arness` section exists or Arness Spark fields are missing, inform the user: "Arness Spark is not configured for this project yet. Run `/arn-brainstorming` to get started — it will set everything up automatically." Do not proceed without it. 4. Create directories if they do not exist
> All references to "Vision directory" and "visual grounding directory" in this skill refer to the configured directories determined above.
**Scaffold (required):** 1. Check the Vision directory for `scaffold-summary.md` 2. If no `## Arness` section found, check `.arness/vision/scaffold-summary.md` at the project root 3. Also check for `package.json` at the project root
**If a scaffold-summary is found:** Read it to extract the full technology stack — UI framework, CSS framework, component library, icon library, dev server command, build command.
**If no scaffold-summary is found:** Inform the user: "The project must be scaffolded before visual sketching. The sketch needs a real CSS framework and component library to generate proposals. Run `/arn-spark-scaffold` first." Do not proceed.
**Architecture vision (required for fallback context):** 1. Check the Vision directory for `architecture-vision.md` 2. If no `## Arness` section found, check `.arness/vision/architecture-vision.md`
**If found:** Read it for UI framework and platform context. The scaffold-summary is the primary source, but the architecture vision provides additional context (platform type, product category).
**If not found:** Proceed if scaffold-summary exists — the scaffold-summary contains enough technical context.
**Product concept (recommended):** 1. Check the Vision directory for `product-concept.md` 2. If no `## Arness` section found, check `.arness/vision/product-concept.md`
**If found:** Read it to identify screens, understand target users, and derive realistic content for sketches.
**If not found:** Ask the user to describe the product briefly. Screen identification and content will be based on the user's description. Note the limitation.
**Dev server verification:** 1. Extract the dev server command from scaffold-summary.md (the "How to Run" section) 2. Verify the project builds without errors by running the build command
If the build fails, do not proceed with sketch generation. Report the error and suggest the user fix the scaffold first.
Workflow
Step 1: Load Context
Read available documents and present a summary:
1. Read `product-concept.md` → extract target users, core experience, product pillars 2. Read `architecture-vision.md` → extract UI framework, platform type 3. Read `scaffold-summary.md` → extract CSS framework, component library, icon library, dev commands 4. Detect framework routing conventions by examining the project source directory:
- Look for `src/routes/` (SvelteKit)
- Look for `app/` with `layout.tsx`/`page.tsx` (Next.js app router)
- Look for `pages/` with `index.tsx` (Next.js pages router) or `index.vue` (Nuxt)
- Fall back to generic approach if none match
Present the context:
"Your project uses **[UI framework]** with **[CSS framework]** and **[component library]** on **[platform]**. The product targets [target users] and focuses on [core experience].
I will create visual direction proposals as real pages in your project, running on your dev server. Each proposal applies a distinct visual approach to the same set of product screens.
Let's start by identifying which screens to sketch."
Step 2: Identify Screens to Sketch
From the product concept (or user description), propose 2-4 key screens that represent the core experience:
"Based on your product concept, I suggest sketching these screens:
1. **[Screen Name]** — [What it shows, e.g., 'Main dashboard with device status grid and connection indicators'] 2. **[Screen Name]** — [What it shows, e.g., 'List view with filterable items and status badges'] 3. **[Screen Name]** — [What it shows, e.g., 'Settings panel with grouped configu
Read more
name: arn-spark-visual-sketch description: >- This skill should be used when the user says "visual sketch", "arn visual sketch", "sketch directions", "explore visuals", "visual proposals", "try different looks", "design directions", "sketch the UI", "visual exploration", "compare styles", "show me options", "what could this look like", or wants to generate multiple visual direction proposals as real HTML/CSS running on the scaffolded project's dev server, iteratively selecting and refining until a final visual direction is chosen. version: 1.0.0
Arness Spark Visual Sketch
Generate visual direction proposals as real HTML/CSS in the scaffolded project, compare them side by side in the browser, and iteratively refine until a direction is selected. This is a conversational skill that runs in normal conversation (NOT plan mode). The primary artifacts are a **visual-direction.md** document in the Vision directory and **screenshot captures** of the selected direction in the visual grounding directory.
This skill sits between `/arn-spark-scaffold` and `/arn-spark-style-explore` in the greenfield pipeline. It produces a visual-direction.md that style-explore uses as primary input for detailed design token specification. The user sees what the product could look like before committing to a full design system.
The iterative process works in rounds: 1. **Round 1:** N proposals with distinct visual directions (default 3) 2. **Round 2+:** N expansions of the selected proposal with user-guided refinements 3. **Final:** The selected direction is captured and documented
Each proposal runs on the scaffold's actual dev server using the project's real CSS framework and component library.
Prerequisites
The following artifacts inform the sketch generation. Check in order:
Determine the output directories: 1. Read the project's `CLAUDE.md` and check for a `## Arness` section 2. If found, extract the configured Vision directory path and Visual grounding directory path 3. If no `## Arness` section exists or Arness Spark fields are missing, inform the user: "Arness Spark is not configured for this project yet. Run `/arn-brainstorming` to get started — it will set everything up automatically." Do not proceed without it. 4. Create directories if they do not exist
> All references to "Vision directory" and "visual grounding directory" in this skill refer to the configured directories determined above.
**Scaffold (required):** 1. Check the Vision directory for `scaffold-summary.md` 2. If no `## Arness` section found, check `.arness/vision/scaffold-summary.md` at the project root 3. Also check for `package.json` at the project root
**If a scaffold-summary is found:** Read it to extract the full technology stack — UI framework, CSS framework, component library, icon library, dev server command, build command.
**If no scaffold-summary is found:** Inform the user: "The project must be scaffolded before visual sketching. The sketch needs a real CSS framework and component library to generate proposals. Run `/arn-spark-scaffold` first." Do not proceed.
**Architecture vision (required for fallback context):** 1. Check the Vision directory for `architecture-vision.md` 2. If no `## Arness` section found, check `.arness/vision/architecture-vision.md`
**If found:** Read it for UI framework and platform context. The scaffold-summary is the primary source, but the architecture vision provides additional context (platform type, product category).
**If not found:** Proceed if scaffold-summary exists — the scaffold-summary contains enough technical context.
**Product concept (recommended):** 1. Check the Vision directory for `product-concept.md` 2. If no `## Arness` section found, check `.arness/vision/product-concept.md`
**If found:** Read it to identify screens, understand target users, and derive realistic content for sketches.
**If not found:** Ask the user to describe the product briefly. Screen identification and content will be based on the user's description. Note the limitation.
**Dev server verification:** 1. Extract the dev server command from scaffold-summary.md (the "How to Run" section) 2. Verify the project builds without errors by running the build command
If the build fails, do not proceed with sketch generation. Report the error and suggest the user fix the scaffold first.
Workflow
Step 1: Load Context
Read available documents and present a summary:
1. Read `product-concept.md` → extract target users, core experience, product pillars 2. Read `architecture-vision.md` → extract UI framework, platform type 3. Read `scaffold-summary.md` → extract CSS framework, component library, icon library, dev commands 4. Detect framework routing conventions by examining the project source directory:
- Look for `src/routes/` (SvelteKit)
- Look for `app/` with `layout.tsx`/`page.tsx` (Next.js app router)
- Look for `pages/` with `index.tsx` (Next.js pages router) or `index.vue` (Nuxt)
- Fall back to generic approach if none match
Present the context:
"Your project uses **[UI framework]** with **[CSS framework]** and **[component library]** on **[platform]**. The product targets [target users] and focuses on [core experience].
I will create visual direction proposals as real pages in your project, running on your dev server. Each proposal applies a distinct visual approach to the same set of product screens.
Let's start by identifying which screens to sketch."
Step 2: Identify Screens to Sketch
From the product concept (or user description), propose 2-4 key screens that represent the core experience:
"Based on your product concept, I suggest sketching these screens:
1. **[Screen Name]** — [What it shows, e.g., 'Main dashboard with device status grid and connection indicators'] 2. **[Screen Name]** — [What it shows, e.g., 'List view with filterable items and status badges'] 3. **[Screen Name]** — [What it shows, e.g., 'Settings panel with grouped configu
Showing the first part of this file.
Arness — H not required. Structured AI workflows for Claude Code. From first idea to production deploy. Seven entry commands. That's all you need to remember.
Repo: AppsVortex/arness
Other skills on arness.
- /arn-assessing
This skill should be used when the user says "assessing", "arness assessing", "assess", "assess codebase", "technical review", "codebase assessment", "find improvements", "what should I improve", "tech debt review", "pattern compliance check", "codebase health check",
Open skill - /arn-code-assess
This skill should be used when the user says "arness code assess", "arn-code-assess", "assess codebase", "technical review", "codebase assessment", "find improvements", "what should I improve", "tech debt review", "tech debt audit", "pattern compliance check", "codebase health
Open skill - /arn-code-batch-cve-fix
This skill should be used when the user says "fix CVEs", "patch vulnerabilities", "apply security patches", "resolve security advisories", "batch CVE fix", "patch dependencies", "fix security findings", "remediate CVEs", "apply CVE fixes", "batch fix vulnerabilities", "resolve
Open skill - /arn-code-batch-cve-scan
This skill should be used when the user says "scan for CVEs", "CVE scan", "check for vulnerabilities", "find vulnerabilities", "check security advisories", "dependabot triage", "dependabot scan", "scan dependencies for security issues", "audit dependencies", "vulnerability
Open skill - /arn-code-batch-implement
This skill should be used when the user says "batch implement", "implement all", "batch execution", "implement all features", "parallel implement", "implement in parallel", "arness batch implement", "arn-code-batch-implement", "run batch implementation", "implement everything",
Open skill - /arn-code-batch-merge
This skill should be used when the user says "batch merge", "merge batch", "arness batch merge", "arn-code-batch-merge", "merge all PRs", "merge batch PRs", "merge the batch", "merge implemented features", "batch merge PRs", "merge open PRs", "merge all feature PRs", "combine
Open skill

