html-email

html-email is a skill for Claude Code from asgeirtj/system_prompts_leaks. It costs 8 tokens per session (623 once invoked), scanned A, original, CC0-1.0.

A single HTML file for an email that is ready to send and designed to work across Gmail, Outlook, and Apple Mail. Email clients are apps that display emails and often support fewer web features than browsers.

In plain words
What is it for?
Use it to create compatible marketing emails, announcements, newsletters, and other styled messages.
Why use it?
It avoids browser features that email clients may remove or display incorrectly, such as JavaScript, external stylesheets, and modern layout systems.

Skill for Claude Code

Written for Claude Code: user-invocable in frontmatter.

Good fit Use it to create compatible marketing emails, announcements, newsletters, and other styled messages.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/asgeirtj/system_prompts_leaks/html-email
About the project

System Prompts Leaks is a collection of captured system instructions used to guide AI chatbots and coding agents before they receive user messages. It serves researchers and developers studying how different AI assistants are directed.

asgeirtj/system_prompts_leaks · 64,614 stars · on GitHub

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 asgeirtj/system_prompts_leaks --skill html-email
Clone the repo
git clone --depth 1 https://github.com/asgeirtj/system_prompts_leaks

Made for: Claude Code.

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.

agentmods badge for html-email

README.md
[![agentmods](https://agentmods.dev/badge/skills/asgeirtj/system_prompts_leaks/html-email/github.svg)](https://agentmods.dev/skills/asgeirtj/system_prompts_leaks/html-email)
Your own site
<a href="https://agentmods.dev/skills/asgeirtj/system_prompts_leaks/html-email"><img src="https://agentmods.dev/badge/skills/asgeirtj/system_prompts_leaks/html-email/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.

agentmods 80×15 button for html-email

Your own site · 80×15
<a href="https://agentmods.dev/skills/asgeirtj/system_prompts_leaks/html-email"><img src="https://agentmods.dev/badge/skills/asgeirtj/system_prompts_leaks/html-email.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 8 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 623 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. ✓ AI security review Fable 5.1 · 6 Sept 2026 📄 Read the review Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00008 $0.00623
Opus 5 $0.00004 $0.00311
Sonnet 5 $0.00002 $0.00125
Haiku 4.5 $0.00001 $0.00062

Measured 11d ago against content hash 9bf6423a7647, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

html-email 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 11d 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.

Origin

Copies of this mod

1 near-identical copy found in the catalogue:

Anthropic/claude-design/skills/html-email/SKILL.md · 55 lines

What it actually says

HTML email

Design an HTML email as ONE self-contained .html file that survives real email clients. Email rendering is not browser rendering — Gmail, Outlook, and Apple Mail each strip or mangle different things, and the rules below are what reliably survives all three. When a rule here conflicts with normal web-design instincts, the rule wins.

Layout and styling:

  • Structure with nested <table role="presentation" cellpadding="0" cellspacing="0" border="0"> — no flexbox, no grid, no floats, no position. One centered wrapper table, max-width 600px, single-column flow (stacked rows beat side-by-side columns).
  • Inline EVERY style on the element it styles. A <style> block in <head> may additionally carry only what can't inline (media queries, dark-mode tweaks) — several clients drop it entirely, so the email must read correctly from inline styles alone.
  • No JavaScript anywhere (universally stripped). No external stylesheets. No web fonts — use email-safe stacks (Arial, Helvetica, Georgia, Verdana, Tahoma, 'Courier New') with generic fallbacks.
  • Build the visual design out of colored table cells, borders, spacer cells, and type — not images. There is nowhere to host project images from here: a referenced project file will not exist for recipients. If imagery is essential, leave a clearly-marked placeholder cell with alt text and tell the user to swap in a hosted https URL before sending.
  • Buttons are "bulletproof": a padded <td> with bgcolor and inline border-radius, the <a> filling it with display:block and inline color — never an image, never a styled <button>.

Client quirks that matter:

  • Outlook (Word engine): give every table/cell explicit widths; set line-height with mso-line-height-rule:exactly; wrap Outlook-only fixes in <!--[if mso]> … <![endif]--> conditionals.
  • Gmail clips messages beyond ~100KB of HTML — stay well under.
  • Add <meta name="color-scheme" content="light dark"> and pick colors that survive dark-mode inversion (avoid pure #000/#fff backgrounds; test text on mid-tone fills).

Deliverability and accessibility:

  • First element in <body>: a hidden preheader span (~85 chars) that previews next to the subject line.
  • alt text on any image, lang on <html>, real <a href> links (no dead # anchors), and a footer with a plausible unsubscribe line and postal address for anything marketing-shaped.

Show the design at 600px; mention in your reply that the file is send-ready HTML the user can drop into their email tool.

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.

  1. 11d ago First seen · 55 lines · 8 tokens per session scan A 9bf6423a7647

Subscribe to this mod's changes

html-email is a skill published in the GitHub repository asgeirtj/system_prompts_leaks (64,614 stars, last pushed yesterday), licensed CC0-1.0. It adds 8 tokens to every session and 623 once invoked, about $0.0000 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.