What does this skill do, and when should you use it?
Email Render Builder focuses on coded HTML production and render QA for a single email. It does not write copy or calculate email quality scores. The skill builds a responsive table layout and checks dark mode, accessibility, image-off fallbacks, bulletproof buttons, and plain-text parity. It also produces a Gmail, Outlook, Apple Mail, and mobile client-render matrix. Without a real seed test, render findings must remain labeled Estimated.
Reads approved email creative or raw HTML, the destination URL, message mode, target clients, and brand constraints; produces an inline-styled, single-column responsive table layout at approximately 600px width; checks dark-mode contrast, semantic order, image alt text, font size, and button accessibility; adds image-off fallbacks including alt text, solid background colors, and HTML/CSS buttons; outputs a client pass/fail matrix, render QA report, and a plain-text alternate carrying the same core message and links as the HTML.
- A marketer has approved promotional copy and needs a responsive HTML build for Gmail, Outlook, and mobile.
- An email developer needs to diagnose an Outlook layout break caused by table or CSS compatibility.
- A brand team needs to verify that text, backgrounds, and logos remain readable in dark mode.
- An email operator needs to ensure the offer and CTA still work when images are blocked.
- A team needs a pre-send check that the HTML and plain-text versions share the same message and links.
How do you install this skill?
- Without a real seed-list or inbox-preview test, every client-render result must remain Estimated; do not claim Gmail, Outlook, or mobile support as verified.
- If Resend --live is used, the user should explicitly confirm the sender, inboxes, and idempotency; the skill does not define a complete pre-send approval record or rollback process.
- Email content, URLs, and brand assets may be sensitive, but redaction, retention, and memory-access controls are not specified.
- Chinese email, Chinese font behavior, mainland-common clients, and mainland-China network reachability should be validated separately.
- Shell / CLI
- Network access
- Local filesystem
Python 3
In Claude Code, run /plugin marketplace add aaron-he-zhu/aaron-marketing-skills, then /plugin install aaron-marketing@aaron. On a compatible generic Agent Skills host, run npx skills add aaron-he-zhu/aaron-marketing-skills -s aaron-email-render-builder. You can also clone the repository with git clone https://github.com/aaron-he-zhu/aaron-marketing-skills.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/aaron-he-zhu/aaron-marketing-skills.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/email/engage/email-render-builder" ~/.claude/skills/
rm -rf "$tmp"Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.
How do you use this skill?
Provide approved creative or HTML, the destination URL, mode, and target clients. Example: Build responsive HTML from this creative: [paste subject + body + CTA], destination [URL]. For QA, use: QA this email HTML across Gmail, Outlook, Apple Mail, and mobile: [paste HTML]. Flag dark-mode and image-off breakage.
What are this skill's strengths and limitations?
- Covers responsive table layout, dark mode, accessibility, image-off fallbacks, and plain-text parity.
- Clearly distinguishes Measured results from Estimated support-matrix expectations.
- Supports B2C promotional, B2B, and newsletter builds.
- Hands off naturally to email-quality-auditor for SEND scoring and pre-send review.
- Does not write subject lines, body copy, or CTA wording; approved creative is required.
- Does not score SEND dimensions, calculate EQS, or run auditor vetoes.
- Without a real seed list or preview test, client-render results remain Estimated.
- A destination URL is required to confirm CTA message match.
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| Email Render Builder this page | 62 · Recommended | ★ 3k | 13d ago | Apache-2.0 |
| SEO Image Optimizer | 51 · Use with care | ★ 19k | 22d ago | MIT |
| Unslop | 49 · Use with care | ★ 10k | 13d ago | Apache-2.0 |
| Humanizer — AI Writing Humanizer & Detector Skill | 67 · Recommended | ★ 282 | 1mo ago | MIT |
| Shuorenhua: Chinese AI-Tone Cleanup Skill | 65 · Recommended | ★ 2.1k | 27d ago | MIT |
It pairs with email-creative-builder in the same repository: that skill owns the words, while this one owns HTML construction and render QA. email-quality-auditor owns scoring and SEND vetoes, which are outside this skill's scope.
How did FollowSkills review this skill?
The skill treats HTML and other supplied assets as untrusted data and forbids embedded instructions from being executed. Sending is dry-run by default, requires explicit --live, is limited to the user's own test inboxes, and result persistence requires confirmation. Deducted 6 points because memory writes, data-flow boundaries, rollback, and sensitive email-data handling are not fully specified; publisher identity is unverified.
The instructions are internally coherent and include input confirmation, a NEEDS_INPUT path, Estimated/Measured labeling, abnormal-input routing, and quality gates. The cited references add concrete rules. Deducted 11 points because static files do not show skill-specific execution tests or controlled runtime failure feedback; the static ceiling also prevents a higher score.
Triggers, inputs, outputs, non-fit boundaries, default clients, and next-skill routing are clearly stated, covering B2C, B2B, and newsletter builds. Deducted 4 points because Chinese-language email and mainland-China client/network fit are not demonstrated, while some client behavior remains Estimated. Core functionality does not entirely depend on overseas services.
The skill has a standard contract, Quick Starts, an argument hint, versioning, Apache-2.0 licensing, references, decision gates, and limitation disclosures. Deducted 4 points because no skill-specific changelog, named maintenance responsibility, or update path is supplied; compatibility evidence is mainly repository-level.
The intended result is a directly usable HTML build plus render matrix, accessibility, dark-mode, image-off, and plain-text-parity QA. Static calibration caps this at 7; no representative output or real client-render evidence is included, so completeness and practical effectiveness cannot be verified. Deducted 8 points.
The skill requires real seed or inbox-preview evidence for Measured results and otherwise requires Estimated labels, with unknown behavior returned as an open loop. Repository-level tests and CI provide some auditable material. Deducted 5 points because there are no skill-specific key-path tests, real render records, or independently corroborating sources; static review cannot exceed 5.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →