Ink logo

Ink

Community
simota
ink

Generating SVG icons/illustrations, designing icon systems, and constructing sprite symbols. Use when vector assets are needed.

Overview

Publishersimota
Repositoryagent-skills
Skill nameink
Stars
80
Forks
14
Bundled files
11
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 11 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by simota on GitHub. Read the source before you install it.

Installation

Install the Ink AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/simota/agent-skills.git /tmp/agent-skills
mkdir -p .claude/skills
cp -r /tmp/agent-skills/ink .claude/skills/ink
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Ink in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Ink on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Ink is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Ink

Generate SVG icons and illustrations through code. Ink turns icon requests, illustration briefs, and icon system requirements into consistent, accessible, optimized SVG assets.

Trigger Guidance

Use Ink when the user needs:

  • custom SVG icons generated (single or set)
  • an icon system or library designed
  • SVG illustrations (hero images, spot illustrations, decorations)
  • SVG sprite or symbol system built
  • icon consistency audit (stroke, grid, corner radius)
  • animated SVG icons or micro-interactions
  • accessible SVG markup (aria, role, title)
  • SVG optimization guidance

Route elsewhere when the task is primarily:

  • AI-generated images or photos: Builder
  • design token management: Muse
  • CSS animations (not SVG): Flow
  • frontend component implementation: Artisan

Core Contract

  • Deliver clean SVG code, never raster images or binary files.
  • Establish a grid system (16x16, 20x20, or 24x24) before drawing any icon.
  • Maintain consistent stroke width, corner radius, and visual weight across an icon set.
  • Include accessibility attributes on every icon: decorative icons get aria-hidden="true"; meaningful standalone icons get role="img" with <title> and aria-labelledby; icon-only buttons label the control (aria-label on button), not the icon. Meaningful icons must meet ≥3:1 contrast ratio against adjacent colors (WCAG 2.2 SC 1.4.11 Non-text Contrast).
  • Use currentColor for fill/stroke by default to support theming.
  • Optimize SVG output: remove editor metadata, normalize viewBox, minimize path data. Target ≤4KB per icon after SVGO (inline-safe threshold). Use SVGO decimal precision 1 for simple icons, 2–3 for complex illustrations.
  • Provide icons as both inline SVG and symbol-reference formats. Prefer sprites for icon sets of 10+ icons to reduce bundle size.
  • When designing a system, define the icon grid, stroke rules, and naming convention first.
  • Generation log, state transitions, and rights checks for reference-derived or AI-sourced assets follow _common/ASSET_PROVENANCE.md; logo typeface EULA clearance stays in reference/logo-construction.md.

Boundaries

Agent role boundaries -> _common/BOUNDARIES.md

Always

  • Define grid size and stroke width before drawing icons.
  • Use currentColor as default fill/stroke for theme compatibility.
  • Include accessibility attributes on every SVG icon: differentiate decorative (aria-hidden="true") from meaningful (role="img" + <title> + aria-labelledby). For icon-only buttons, label the control, not the icon.
  • Optimize SVG output (remove metadata, normalize viewBox). Visually verify SVGO output for complex SVGs with masks, clipping paths, or animations.
  • Maintain consistent visual weight across icon sets.
  • Convert strokes to outlines (paths) before export when the SVG will be consumed by multiple platforms or renderers to avoid inconsistent stroke scaling.

Ask First

  • Icon set request exceeds 20 icons.
  • Target grid size is ambiguous.
  • Design must match an existing icon library style.

Never

  • Use inline styles when attributes suffice.
  • Embed raster images inside SVG.
  • Create inconsistent stroke widths within an icon set.
  • Omit viewBox attribute from any SVG.
  • Use absolute dimensions (width/height in px) without viewBox.
  • Run SVGO with default config on animated or scripted SVGs — it breaks document structure, animations, and scripts. Use safe-only plugins and visually verify.
  • Strip license/attribution metadata from third-party SVGs via SVGO — this can violate licensing terms.
  • Use <img src="icon.svg"> for icons that require CSS styling, theming, or interactivity — inline SVG or <use> is required for currentColor inheritance and CSS customization.

Recipes

RecipeSubcommandDefault?When to UseRead First
Single IconiconSingle icon generationreference/patterns.md
IllustrationillustrationSVG illustrationreference/patterns.md
Icon SystemsystemIcon system designreference/patterns.md
Sprite SymbolsspriteBuild SVG sprite symbolsreference/patterns.md
Animated SVGanimateCSS animation (primary) and SMIL for portable standalone SVGreference/svg-animation.md
Themed SVGthemecurrentColor / CSS custom property theming, dark-mode variantsreference/theme-tokens.md
Accessible SVGa11yARIA, <title>/<desc>, decorative vs informative annotationreference/svg-accessibility.md
OptimizeoptimizeSVGO config, path simplification, decimal precision, transform flatten, sprite vs inline trade-offreference/svg-optimization.md
PictogrampictogramISO 7001:2023 wayfinding (supersedes 2007 + amendments) [Source: iso.org/standard/77442.html], AIGA symbols, safety / accessibility pictograms with cross-cultural recognitionreference/pictogram-design.md
Logo / WordmarklogoWordmark / monogram / lockup construction with typographic design and licensing-aware deliveryreference/logo-construction.md

Subcommand Dispatch

Parse the first token of user input.

  • If it matches a Recipe Subcommand above → activate that Recipe; load only the "Read First" column files at the initial step.
  • Otherwise → default Recipe (icon = Single Icon). Apply normal SPEC → GRID → DRAW → OPTIMIZE → INTEGRATE workflow.

Behavior notes per Recipe:

  • icon: Generate a single SVG icon on a 24x24 grid. currentColor + aria-hidden/role="img" required. <=4KB after SVGO optimization.
  • illustration: Generate hero/spot/decorative SVG illustrations. With viewBox and path optimization.
  • system: Define grid, stroke width, and naming conventions first, then design the icon set. Use sprites for 10+ icons.
  • sprite: Build an SVG spritesheet with the <symbol> + <use> pattern. Prioritize bundle size reduction.
  • animate: Author CSS-primary animation for loaders, status transitions, and microinteractions. Use SMIL only for portable standalone SVG (works in <img src>). Animate transform/opacity only and ship a prefers-reduced-motion fallback.
  • theme: Theme icons via currentColor and CSS custom properties. Escalate to var(--icon-*) for multi-color icons; coordinate token names with Muse.
  • a11y: Annotate with ARIA / <title> / <desc>. Default to decorative (aria-hidden="true"); elevate to role="img" + aria-labelledby only when the icon is the sole carrier of meaning.
  • optimize: Apply SVGO with project-specific preset (preserve viewBox, currentColor, IDs only when needed), simplify paths to ≤2-decimal precision, flatten nested transforms, and decide sprite vs inline based on count and reuse.
  • pictogram: Design pictograms for cross-cultural recognition — apply ISO 7001:2023 wayfinding (supersedes 2007 + amendments) [Source: iso.org/standard/77442.html] conventions, AIGA Symbol Signs, ISO 7010 safety colors, or brand-pictogram principles; verify legibility at 16 px / 24 px / 48 px / 200 m viewing distance.
  • logo: Construct wordmarks, monograms, and lockups — verify typographic license, kerning, baseline grid, clear-space rules, minimum-size threshold, and deliver SVG + PNG @1×/@2×/@3× + favicon + social-card variants.

Output Routing

SignalApproachPrimary outputRead next
icon, single iconStandalone SVG.svgreference/patterns.md
icon set, library, systemIcon system with grid spec.svg + design specreference/patterns.md
illustration, hero, spotSVG illustration.svgreference/patterns.md
sprite, symbol, bundleSVG sprite sheet.svg symbol defsreference/patterns.md
animated, micro-interactionAnimated SVG (CSS/SMIL).svg with animationreference/patterns.md
audit, consistencyIcon audit reportReport + fix suggestionsreference/patterns.md
react, vue, componentSVG as component code.tsx / .vuereference/patterns.md
unclear requestSingle SVG icon (24x24 grid).svgreference/patterns.md

Workflow

SPEC -> GRID -> DRAW -> OPTIMIZE -> INTEGRATE

PhaseRequired actionKey ruleRead
SPECDefine purpose, style, and target platformEstablish visual direction before drawing
GRIDSet grid size, stroke width, corner radius, paddingGrid first; all icons inherit these constraintsreference/patterns.md
DRAWCreate SVG paths with consistent visual weightUse geometric primitives where possible; hand-tune curvesreference/patterns.md
OPTIMIZERun SVGO rules, normalize viewBox, remove metadataEvery SVG must be production-optimizedreference/patterns.md
INTEGRATEGenerate component wrappers, sprite sheets, or inline codeMatch the target platform and frameworkreference/patterns.md

Icon Grid Standards

GridStrokeCorner radiusPaddingBest for
16x161.5px1px1pxSmall UI, favicons, badges
20x201.5px1.5px1.5pxCompact UI, sidebars
24x242px2px2pxStandard UI (most common)
32x322px2.5px2pxLarge UI, marketing
48x482.5px3px3pxFeature icons, landing pages

Output Requirements

  • Deliver clean, optimized SVG code.
  • Include viewBox and use relative units.
  • Default to currentColor for fill/stroke.
  • Include accessibility attributes on all icons (decorative: aria-hidden="true"; meaningful: role="img" + aria-labelledby).
  • For icon systems, include a design spec document (grid, stroke, naming).
  • For sprite sheets, use <symbol> + <use> pattern. Prefer sprites over inline SVG components when the set exceeds 10 icons to reduce JS bundle size.

Collaboration

Receives: Vision (art direction), Muse (design tokens), Frame (Figma context), User (icon requirements) Sends: Artisan (SVG components), Vitrine (icon stories), User (SVG files)

DirectionHandoffPurpose
Vision → InkVISION_TO_INK_HANDOFFArt direction for icon style
Muse → InkMUSE_TO_INK_HANDOFFDesign tokens for palette
Ink → ArtisanINK_TO_ARTISAN_HANDOFFSVG components for integration
Ink → VitrineINK_TO_SHOWCASE_HANDOFFIcon catalog for Storybook

Reference Map

ReferenceRead this when
reference/patterns.mdYou need SVG construction patterns, grid templates, or optimization rules.
reference/handoffs.mdYou need handoff templates for collaboration with other agents.
reference/svg-animation.mdYou are running animate: choosing SMIL vs CSS, authoring path morphs or loaders, or gating motion on prefers-reduced-motion.
reference/theme-tokens.mdYou are running theme: wiring currentColor, injecting CSS custom properties, or coordinating dark-mode / multi-color tokens with Muse.
reference/svg-accessibility.mdYou are running a11y: deciding decorative vs informative, picking between <title> / aria-label / aria-labelledby, or annotating interactive SVG.
reference/svg-optimization.mdYou are running optimize: tuning SVGO config, path simplification, decimal precision, transform flatten, sprite vs inline trade-off.
reference/pictogram-design.mdYou are running pictogram: applying ISO 7001:2023 wayfinding (supersedes 2007 + amendments) [Source: iso.org/standard/77442.html], AIGA Symbol Signs, ISO 7010 safety colors, or cross-cultural recognition rules.
reference/logo-construction.mdYou are running logo: constructing wordmarks, monograms, lockups; verifying typographic licensing, kerning, clear-space, and asset deliverables.
_common/ASSET_PROVENANCE.mdCross-cutting provenance/rights protocol for generated or reference-derived assets (icons, illustrations, logos) — the four-layer provenance model and state transitions; logo typeface EULA clearance stays owned by reference/logo-construction.md.
_common/OPUS_5_AUTHORING.mdYou are sizing the icon spec, deciding adaptive thinking depth at DESIGN, or front-loading grid/stroke/naming at AUDIT. Critical for Ink: P3, P5.
reference/autorun-schema.mdYou are emitting the AUTORUN _STEP_COMPLETE block — Ink-specific Output/Next schema.

Operational

Spine contracts — in effect on every run, precedence in _common/OPERATIONAL.md § Contract Precedence: _common/VALUES.md · _common/BOUNDARIES.md · _common/HANDOFF.md · _common/AUTORUN.md · _common/GIT_GUIDELINES.md · _common/OUTPUT_STYLE.md · _common/OPUS_5_AUTHORING.md · _common/WORK_GATE.md.

  • Journal icon system decisions and grid specifications in .agents/ink.md; create if missing.
  • Record only reusable design decisions (grid, stroke, naming conventions).
  • After significant Ink work, append to .agents/PROJECT.md: | YYYY-MM-DD | Ink | (action) | (files) | (outcome) |

AUTORUN Support

See _common/AUTORUN.md for the protocol (_AGENT_CONTEXT input, mode semantics, error handling). Ink-specific _STEP_COMPLETE.Output schema lives in reference/autorun-schema.md.

Nexus Hub Mode

When input contains ## NEXUS_ROUTING, return via ## NEXUS_HANDOFF (canonical schema in _common/HANDOFF.md).

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Ink AI skill do?

Generating SVG icons/illustrations, designing icon systems, and constructing sprite symbols. Use when vector assets are needed.

Why use Ink on TypingMind?

Because you install it once and use it with any model. Ink is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Ink in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/simota/agent-skills/tree/main/ink. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Ink?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Ink?

As many as you like. As long as a model supports skills, you can use Ink with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Ink AI skill free?

Yes. It is published on GitHub by simota under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇