A tool for creating diagrams from a short text-based description, then exporting them as SVG or PNG images. SVG is a scalable image format that stays sharp when resized.

Use it for flowcharts, charts, and simple two-dimensional diagrams with colors selected for a chosen brand.

It avoids drawing every box, arrow, and label by hand and can check the result for overlaps, overflow, and label collisions.

svg — Skill for Claude Code by marsmike/feinschmiede

Install

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.

Any agent

npx skills add marsmike/feinschmiede --skill svg

Clone the repo

git clone --depth 1 https://github.com/marsmike/feinschmiede

Made for: Claude Code.

Or install feinbild, the plugin that ships this one along with the rest of its 3 skills, 3 commands.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/marsmike/feinschmiede/svg/github.svg)](https://agentmods.dev/skills/marsmike/feinschmiede/svg)

Your own site

<a href="https://agentmods.dev/skills/marsmike/feinschmiede/svg"><img src="https://agentmods.dev/badge/skills/marsmike/feinschmiede/svg/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.

Your own site · 80×15

<a href="https://agentmods.dev/skills/marsmike/feinschmiede/svg"><img src="https://agentmods.dev/badge/skills/marsmike/feinschmiede/svg.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 28 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 502 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00028	$0.00502
Opus 5.5	$0.00011	$0.00201
Sonnet 5.5	$0.00006	$0.00100
Haiku 4.5	$0.00003	$0.00050

Measured 1mo ago against content hash 663599c76c36, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

svg 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 1mo 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.

feinbild/skills/svg/SKILL.md · 46 lines

What it actually says

feinbild — SVG diagrams

Quick Start

/svg diagram.svg.dsl

feinbild is a command on your PATH. Two steps: expand a .svg.dsl to .svg (brand colors resolved), then render to PNG.

feinbild svg expand chart.svg.dsl --brand feinschliff   # -> chart.svg
feinbild svg render chart.svg                           # -> chart.png
feinbild verify chart.svg                               # lint: overflow, overlap, label collision

--brand (or a leading @brand <name> line in the DSL) selects the brand; render takes no brand (it consumes already-resolved colors). Write outputs into the project so other plugins can consume them.

Default to clean

≤3 arrows per flow. More than that means the layout is wrong — re-chunk or pick another visualization.
No text over shapes. Labels live inside boxes or adjacent via a connector — never on a border.
Aspect-aware. Default 16:9 (1920×1080). If the brief specifies a target slot match it; otherwise embeds get squished.
Metaphor, not clutter. Briefs saying "messy" / "tangled" / "stitched" describe a CONCEPT — convey via composition (overlap, dim color, tight grouping), not by drawing actual chaos.

References

DSL reference — canvas header, basic + extended primitives, the 17-name semantic color vocabulary, the virtual: viewport.
Examples — small / medium charts (bar chart, stat card grid, annotated chart).
Deep examples — dense compositions with the extended primitive set (stacked_bar, brace, callout, swatch_grid, label_box, polyline, path); see examples/yocto-build-pipeline.svg.dsl.

Workflow: author a .svg.dsl, run feinbild svg expand then render, inspect the PNG, and do at most one DSL fix loop before escalating to the references above.

Files

What ships with it

5 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.

Changes

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.

1mo ago First seen · 46 lines · 28 tokens per session scan A 663599c76c36

Subscribe to this mod's changes

svg is a skill published in the GitHub repository marsmike/feinschmiede (3 stars, last pushed 1mo ago), licensed MIT. It adds 28 tokens to every session and 502 once invoked, about $0.0001 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 2026-08-31.

Claude Code Design and UX agent-skills agentic-coding anthropic brand-system claude

mobile-ios-design

Master iOS Human Interface Guidelines (HIG) and SwiftUI patterns to build polished, native iOS applications that feel at home on Apple platforms.

HigorAlves/orc · 16 tokens

tailwind-design-system

Build scalable Tailwind CSS v4 design systems — design tokens, component libraries, responsive patterns.

HigorAlves/orc · 23 tokens

creo-design-review

Comprehensive UI/UX design review with Playwright. Evaluates responsive design (375-1920px), WCAG AA accessibility, Nielsen's 10 heuristics, component consistency, and visual polish. Trigger keywords: design review, UI review, accessibility audit, responsive check, heuristic evaluation.

oyusypenko/creo · 64 tokens

creo-ux-competitor

Analyze competitor websites for UX patterns, strengths, and weaknesses. Navigates external URLs, documents UX patterns, evaluates against Nielsen heuristics, and extracts actionable learnings. Trigger keywords: competitor analysis, competitor UX, competitive analysis, benchmark, compare competitors.

oyusypenko/creo · 58 tokens

landing-not-slop

Marketing pages that do not read as machine-generated. Use when writing or reviewing a landing, hero, pricing table, feature grid or marketing copy.

avanturer/claude-bestpractice · 34 tokens

republic-design

Republic-style UX design reviewer and scorer for pitch decks, landing pages, and investor-facing materials. Applies Republic.com's design DNA (financial-grade trust, data-forward, restrained elegance) to score and improve any visual artifact. Use when user says "republic", "republic style", "score design", "polish…

floomhq/moto · 86 tokens