Trmnl logo

Trmnl

Community
viralcode
trmnl

Generate content for TRMNL e-ink display devices using the TRMNL CSS framework. Use when the user wants to display information on their TRMNL device, send messages to an e-ink display, create dashboard content, show notifications, or update their terminal display. Supports rich layouts with the TRMNL framework (flexbox, grid, tables, progress bars, typography utilities) and sends content via webhook API.

Overview

Publisherviralcode
Repositoryopenwhale
Skill nametrmnl
Stars
65
Forks
11
Bundled files
14
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.

  • 14 bundled files

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

  • Open source

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

Installation

Install the Trmnl 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/viralcode/openwhale.git /tmp/openwhale
mkdir -p .claude/skills
cp -r /tmp/openwhale/skills/trmnl .claude/skills/trmnl
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Trmnl 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 Trmnl 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 Trmnl 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.

TRMNL Content Generator

Generate HTML content for TRMNL e-ink display devices.

Quick Start Workflow

  1. Check for $TRMNL_WEBHOOK environment variable
  2. If missing, prompt user for webhook URL
  3. Ask user to verify TRMNL display markup is set to: <div>{{content}}</div>
  4. Confirm device type (default: TRMNL OG, 2-bit, 800x480)
  5. Read relevant reference docs based on content needs
  6. Generate HTML using TRMNL framework classes
  7. Send via POST to webhook (use temp file method)
  8. Send minimal confirmation only - Do NOT echo content back to chat

Device & Setup

Default target: TRMNL OG (7.5" e-ink, 800x480px, 2-bit)

Display markup required:

html
<div>{{content}}</div>

Webhook:

bash
export TRMNL_WEBHOOK="https://trmnl.com/api/custom_plugins/{uuid}"

Sending content (temp file method):

bash
cat > /tmp/trmnl.json << 'EOF'
{"merge_variables":{"content":"<div class=\"layout\">HTML</div>"}}
EOF
curl "$TRMNL_WEBHOOK" -H "Content-Type: application/json" -d @/tmp/trmnl.json -X POST

Webhook Limits

TierPayload SizeRate Limit
Free2 KB (2,048 bytes)12 requests/hour
TRMNL+5 KB (5,120 bytes)30 requests/hour

Check payload size before sending:

bash
python3 scripts/check_payload.py /tmp/trmnl.json

Tips to reduce payload size:

  • Minify HTML (remove unnecessary whitespace)
  • Use framework classes instead of inline styles
  • Use short class names the framework provides
  • Remove comments from HTML
  • Use deep_merge strategy for incremental updates

Reference Documentation

Read these files as needed:

FileWhen to Read
references/patterns.mdStart here - Common plugin patterns from official examples
references/framework-overview.mdDevice specs, e-ink constraints, responsive prefixes
references/css-utilities.mdColors, typography, sizing, spacing utilities
references/layout-systems.mdFlexbox, grid, overflow/clamp engines
references/components.mdTitle bar, dividers, items, tables, charts
references/webhook-api.mdPayload format, rate limits, troubleshooting
assets/anti-patterns.mdCommon mistakes to avoid
assets/good-examples/HTML reference implementations

Scripts:

ScriptPurpose
scripts/check_payload.pyVerify payload size before sending (run on /tmp/trmnl.json)

Standard Plugin Structure

Every plugin follows this pattern:

html
<div class="layout layout--col gap--space-between">
  <!-- Content sections separated by dividers -->
</div>
<div class="title_bar">
  <img class="image" src="icon.svg">
  <span class="title">Plugin Name</span>
  <span class="instance">Context</span>
</div>
  • layout + layout--col = vertical flex container
  • gap--space-between = push sections to edges
  • title_bar = always at bottom, outside layout
  • divider = separate major sections
  • CRITICAL: Only ONE .layout element per view (no nesting)

Grid System (10-Column)

Column spans should sum to 10:

html
<div class="grid">
  <div class="col--span-3">30%</div>
  <div class="col--span-7">70%</div>
</div>

Simple equal columns: grid--cols-2, grid--cols-3, etc.

Item Component

Standard data display pattern:

html
<div class="item">
  <div class="meta"><span class="index">1</span></div>
  <div class="content">
    <span class="value value--xlarge value--tnums">$159,022</span>
    <span class="label">Total Sales</span>
  </div>
</div>

Value Typography

Always use value--tnums for numbers.

ClassUsage
value--xxxlargeHero KPIs
value--xxlargeLarge prices
value--xlargeSecondary metrics
value--smallTertiary data
value--tnumsAlways for numbers

Auto-fit: <span class="value" data-fit-value="true">...</span>

Columns (for Lists)

html
<div class="columns">
  <div class="column" data-overflow="true" data-overflow-counter="true">
    <span class="label label--medium group-header">Section</span>
    <div class="item">...</div>
  </div>
</div>

Grayscale Dithering

Use dithered classes, not inline gray colors:

  • bg--black, bg--gray-60, bg--gray-30, bg--gray-10, bg--white
  • text--black, text--gray-50

Data Attributes

AttributePurpose
data-fit-value="true"Auto-resize text to fit
data-value-format="true"Auto-format numbers (locale-aware)
data-clamp="N"Limit to N lines
data-overflow="true"Enable overflow management
data-overflow-counter="true"Show "and X more"
data-overflow-max-cols="N"Max columns for overflow
data-content-limiter="true"Auto-adjust text size
data-pixel-perfect="true"Crisp text rendering
data-table-limit="true"Table overflow with "and X more"

Label & Title Variants

html
<span class="label label--small">Small</span>
<span class="label label--medium">Medium</span>
<span class="label label--underline">Underlined</span>
<span class="label label--gray">Muted/completed</span>
<span class="title title--small">Compact title</span>

Gap Utilities

gap--space-between | gap--xxlarge | gap--xlarge | gap--large | gap--medium | gap | gap--small | gap--xsmall | gap--none

Typography Guidelines

Recommended: Georgia serif font for e-ink readability.

Content-aware sizing:

  • Short content = bigger fonts (24-28px body)
  • Long content = smaller fonts (20px body)
  • Headings: 36-48px

User Experience

Critical: Do NOT echo content back to chat. Just confirm "Sent to TRMNL".

Anti-Patterns

  1. Tiny fonts for short content
  2. Center-aligning columns with different lengths (use layout--start)
  3. Spoiling content in chat confirmation
  4. Missing value--tnums on numbers
  5. Missing title_bar
  6. Not using data-fit-value on primary metrics
  7. Skipping data-overflow on variable lists
  8. Using inline gray colors instead of bg--gray-*
  9. Forgetting dividers between sections

Best Practices

  1. Verify <div>{{content}}</div> display markup
  2. Use layout + title_bar structure
  3. Always value--tnums for numbers
  4. Use data-fit-value on primary metrics
  5. Use data-overflow on variable lists
  6. Use item component pattern
  7. Use divider between sections
  8. Use bg--gray-* dithered classes
  9. Content-aware font sizing
  10. Top-align columns (layout--start)
  11. Temp file method for curl
  12. Minimal confirmations

Mashup Layouts (Multi-Plugin)

For dashboard views with multiple plugins:

LayoutDescription
mashup--1Lx1R2 columns (50/50)
mashup--1Tx1B2 rows (50/50)
mashup--2x24 quadrants

See references/layout-systems.md for all 7 layouts.

Troubleshooting

ProblemSolution
Webhook failsVerify URL, check rate limits (12/hour free)
Content missingCheck display markup is <div>{{content}}</div>
Payload too largeRun scripts/check_payload.py, keep under 2KB (free) or 5KB (TRMNL+)
Numbers misalignedAdd value--tnums
Text overflowUse data-clamp or data-overflow
Columns misalignedUse layout--start not layout--center
Multiple layouts errorKeep only ONE .layout element per view
Nested content failsUse .richtext for nested/formatted content

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 Trmnl AI skill do?

Generate content for TRMNL e-ink display devices using the TRMNL CSS framework. Use when the user wants to display information on their TRMNL device, send messages to an e-ink display, create dashboard content, show notifications, or update their terminal display. Supports rich layouts with the TRMNL framework (flexbox, grid, tables, progress bars, typography utilities) and sends content via webhook API.

Why use Trmnl on TypingMind?

Because you install it once and use it with any model. Trmnl 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 Trmnl in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/viralcode/openwhale/tree/main/skills/trmnl. 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 Trmnl?

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 Trmnl?

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

Is the Trmnl AI skill free?

It is published on GitHub by viralcode. Check the repository for licensing terms. 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 👇