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 agentmods add skills/firstp1ck/pi-coding-agent-forge/desktop-visual-designnpx skills add Firstp1ck/pi-coding-agent-forge --skill desktop-visual-designgit clone --depth 1 https://github.com/Firstp1ck/pi-coding-agent-forgeWrote 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/firstp1ck/pi-coding-agent-forge/desktop-visual-design)<a href="https://agentmods.dev/skills/firstp1ck/pi-coding-agent-forge/desktop-visual-design"><img src="https://agentmods.dev/badge/skills/firstp1ck/pi-coding-agent-forge/desktop-visual-design.svg" alt="Measured on agentmods" 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 | $0.00071 | $0.02674 |
| Opus 5 | $0.00036 | $0.01337 |
| Sonnet 5 | $0.00014 | $0.00535 |
| Haiku 4.5 | $0.00007 | $0.00267 |
Grade A, and why
desktop-visual-design 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 4d 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 — 131 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Desktop Visual Design
Build system-native desktop interfaces from the visual rules already present in the user's environment. Inspect the target first, turn repeated values into semantic tokens, reuse host components, and verify the result across interaction states and theme changes. When the target has no visual contract, use the restrained terminal-first fallback defined here rather than inventing a glossy web-app style.
Installing this Pi package makes the skill available to Pi. It does not modify desktop settings, themes, or application code by itself.
When to use
Trigger for:
- Quickshell widgets, panels, bars, popups, overlays, launchers, and notifications.
- Desktop apps built with QML, Qt, GTK, Tauri, Electron, or embedded web views.
- UI reviews focused on palette, hierarchy, spacing, typography, borders, geometry, focus states, and motion.
- Requests to make a desktop interface match an existing system theme or first-party surface.
Do not trigger for server-side work, CLI output formatting, compositor configuration unrelated to application appearance, or general public-website design without a desktop-integration requirement.
Principles
- Inspect before styling. Existing theme files, token objects, shared controls, and first-party surfaces are the source of truth.
- Honor the active color scheme. Standalone apps must read the toolkit, portal, or host-shell light or dark preference before choosing a built-in palette. For Linux Qt apps, prefer an explicit XDG portal result at startup because Qt platform-theme environment can differ between launchers; use Qt style hints as the fallback and live-change input. Automatic mode must update when its authoritative preference changes.
- Use semantic roles. Components depend on foreground, background, accent, urgent, muted, and named surface roles instead of literal palette colors.
- Define every interaction state. Cover idle, hover, keyboard cursor, focus, selected, pressed, disabled, and urgent states where they apply.
- Keep geometry stable. Reserve border and focus-ring space so interaction never shifts layout.
- Use one scale. Derive spacing and typography from named tokens and a documented base size.
- Retheme through one path. Parse, validate, and publish an atomic theme snapshot. Theme changes must not leave mixed old and new values on screen.
- Match existing motion. Honor the target's motion language and reduced-motion settings. When no motion language exists, animate only short color or opacity changes.
- Restrain decoration. Without an explicit reference, prefer opaque flat surfaces, one accent family, thin borders, square geometry, and no gradients, glow, blur, hover lift, or decorative shadows.
- Compare relationships, not sampled pixels. A screenshot can establish palette relationships, hierarchy, composition, framing, type treatment, landmarks, and state language. It does not override project tokens with guessed colors or justify forcing its apparent color scheme.
What ships with it
3 files 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.
- 4d ago First seen · 131 lines · 71 tokens per session scan A 91667bf4376d
desktop-visual-design is a skill published in the GitHub repository Firstp1ck/pi-coding-agent-forge (74 stars, last pushed 4d ago), licensed MIT. It adds 71 tokens to every session and 2,674 once invoked, about $0.0004 per session on Opus 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 2026-08-30.
Other skills, from other repositories
frontend-philosophy
Visual & UI philosophy (The 5 Pillars of Intentional UI). Understand deeply to avoid "AI slop" and create distinctive, memorable interfaces.
Mandu Styling
Tailwind CSS v4 integration and Island styling patterns for Mandu Framework.
diceui
Build complex, accessible UI components using DiceUI — combobox, tags-input, sortable, kanban, data-table, file-upload, color-picker, and 40+ more. Use when the user wants to add any DiceUI component, mentions @diceui packages, asks for complex interactive components like combobox, sortable lists, kanban boards…
github-release-management
Comprehensive GitHub release orchestration with AI swarm coordination for automated versioning, testing, deployment, and rollback management.
flow-nexus-platform
Comprehensive Flow Nexus platform management - authentication, sandboxes, app deployment, payments, and challenges.
github-code-review
Comprehensive GitHub code review with AI-powered swarm coordination.