Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add build-with-dhiraj/ai-workflow-framework-portability-kit --skill claude-design-promptinggit clone --depth 1 https://github.com/build-with-dhiraj/ai-workflow-framework-portability-kitWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/build-with-dhiraj/ai-workflow-framework-portability-kit/claude-design-prompting)<a href="https://agentmods.dev/skills/build-with-dhiraj/ai-workflow-framework-portability-kit/claude-design-prompting"><img src="https://agentmods.dev/badge/skills/build-with-dhiraj/ai-workflow-framework-portability-kit/claude-design-prompting/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/build-with-dhiraj/ai-workflow-framework-portability-kit/claude-design-prompting"><img src="https://agentmods.dev/badge/skills/build-with-dhiraj/ai-workflow-framework-portability-kit/claude-design-prompting.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00102 | $0.01409 |
| Opus 5.5 | $0.00041 | $0.00564 |
| Sonnet 5.5 | $0.00020 | $0.00282 |
| Haiku 4.5 | $0.00010 | $0.00141 |
Grade A, and why
claude-design-prompting scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 25d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Claude Design prompt writing
Claude Design builds a first version you react to, then you refine in rounds. Prompts that fight this model (one giant do-everything prompt) produce worse results than prompts that work with it. Everything below is grounded in Anthropic's prompting best practices, the Claude Design support docs and tutorials, and the frontend-aesthetics cookbook.
Pick the right channel first
Claude Design has three editing channels. Choosing wrong wastes rounds:
- Chat prompt: structural or repeated changes (new sections, status models, table columns, a pattern applied across the screen). This skill is about these.
- Inline comment on an element: one-element nits (padding, a label, one button). Don't write a chat prompt for these. Known issue: inline comments occasionally fail to persist; re-check after applying.
- Canvas editor: drag, resize, align. No prompt needed.
The edit-prompt formula
One concern per prompt, delivered as a sequence. Each prompt:
- Scope line first. Name the screen or region being edited, and what must not change. Claude Design edits eagerly; an explicit freeze list is the single highest-value sentence. Example: "Edit the All Labs table only. Leave the modals, tabs, and toolbar unchanged."
- Numbered, ordered instructions. Sequential numbered steps, one visible change each, naming the element as the user sees it ("the Created by column", "the caption under the toolbar", "slide 3"). Never describe changes by code or file location; describe what is visible.
- State the why when it constrains the solution. Claude generalizes from motivation: "the team manages 1,000+ accounts, so no plain scroll list" gets a type-ahead everywhere it matters; a bare "add search to the filter" gets one search box.
- Exact copy in quotes. Any label, caption, or microcopy the design must show is given verbatim in quotes, not paraphrased.
- Sample data is part of the design. If the change implies different data (a new status, a different persona in a column), say what the sample rows should now show. Claude Design edits data and layout together.
- Ban words explicitly, with the reason. "Never use the word Published anywhere; it is reserved for article publishing" outperforms hoping it won't appear.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 25d ago First seen · 61 lines · 102 tokens per session scan A 3bb36123c956
claude-design-prompting is a skill published in the GitHub repository build-with-dhiraj/ai-workflow-framework-portability-kit (5 stars, last pushed 12d ago), licensed MIT. It adds 102 tokens to every session and 1,409 once invoked, about $0.0004 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-21.
Other skills, from other repositories
prototype
Build a throwaway prototype to answer a design question. Use when the user wants to sanity-check whether a state model or logic feels right, or explore what a UI should look like.
anatomy
Explain an idea as an interactive isometric machine figure.
qa-test-planner
Generate comprehensive test plans, manual test cases, regression test suites, and bug reports for QA engineers. Includes Figma MCP integration for design validation.
ux-design
Section-by-section UX spec authoring for a screen, flow or HUD. Reads the player journey to provide context; also project-wide accessibility.
visual-artifact-design
A general workflow for creating, editing, diagnosing, or checking things people will see, read, or use. It defines the intended medium, audience, required outcome, evidence, and final consumer before work begins.
visualize
Render a polished visual inline in the chat as part of your answer — a diagram, a chart, an interactive explainer, or a UI mockup. Load it proactively whenever an explanation would land better as a picture than as prose. Do not wait to be asked.