What does this skill do, and when should you use it?
This skill belongs to the Email/SEND Engage phase of Aaron Marketing Skills and adds a personalization layer to already-written email creative. It maps merge tags to real export columns, defines natural rendered fallbacks, and configures conditional content for named audience segments. It also audits empty fields, dead branches, and PII exposure, while noting relevance to SEND’s E dimension for engagement and personalization. It does not build segments, write base email copy, or calculate EQS.
Reads the existing email creative, a segment map or export-column list, the sending mode, and field fill-rates. It produces a merge-tag map with each source column, field type, explicit fallback, and rendered fallback; conditional if/elseif/else rules tied to named segments with a mandatory catch-all else; a worst-case fallback-safety audit; and a PII guard covering fields allowed or blocked in the body and subject line. It checks personalized promotional claims against the claims ledger and flags unsupported wording as needs source. The output is a four-part personalization spec plus a standard handoff summary, with optional saving to the specified memory path after user confirmation.
- A lifecycle marketer has finished an email and needs first_name, city, or product-category personalization without producing renders such as “Hi ,”.
- A CRM team has named segments such as champions and at-risk and needs different content blocks while retaining a valid default branch.
- A B2B outbound team wants personalization grounded in existing, verifiable export fields rather than guessed contact details.
- An email operator needs a pre-send review of low-fill fields, broken conditionals, empty renders, and PII exposure.
- A newsletter team needs safe defaults and segment-specific variations for dynamic editorial or promotional modules.
How do you install this skill?
- Before use, provide the existing email creative, named segments or an export-column list, and fill rates for consequential fields; do not invent tags, segments, or copy when these are missing.
- Saving and promoting to memory affects host runtime state; confirm that the host supports the paths, permissions, and failure recovery, then use the render builder and quality auditor before sending.
- The PII guard is a documented instruction, not a verified rendering enforcement mechanism; ESP-specific merge-tag and conditional syntax still requires platform testing.
- The publisher is not verified by the FollowSkills enterprise registry and should be treated as unknown provenance.
- Local filesystem
The skill is distributed in the Apache-2.0-licensed repository aaron-he-zhu/aaron-marketing-skills. In Claude Code, run /plugin marketplace add aaron-he-zhu/aaron-marketing-skills, then /plugin install aaron-marketing@aaron. On other Agent Skills-compatible hosts, run npx skills add aaron-he-zhu/aaron-marketing-skills, or clone the repository with git clone https://github.com/aaron-he-zhu/aaron-marketing-skills. The skill file is email/engage/dynamic-content-personalizer/SKILL.md.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/aaron-he-zhu/aaron-marketing-skills.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/email/engage/dynamic-content-personalizer" ~/.claude/skills/
rm -rf "$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?
Once installed, send your agent any of these to trigger it:
- Add merge tags with fallbacks to this email; export columns are first_name, city, last_product; promo mode.
Provide an existing email creative, a segment map or export-column list, field fill-rates, and one mode: promo, cold, or newsletter. Example: “Add merge tags with fallbacks to this email; export columns are first_name, city, last_product; promo mode.” If the creative or segment/column inputs are missing, the skill asks for them. If a conditional offer depends on an unknown fill-rate, it asks for the rate or defaults the decision to the catch-all branch.
What are this skill's strengths and limitations?
- Requires an explicit, natural fallback for every merge tag.
- Requires named segment conditions and a catch-all else branch.
- Audits empty renders, dead conditionals, low-fill fields, and PII exposure.
- Works from user-provided exports and does not require a specific ESP or MCP.
- Defines clear handoffs to email-render-builder and email-quality-auditor.
- Does not build segments, write the base email, or score EQS.
- Reliable field binding and conditional coverage require actual column names and fill-rates.
- The source does not document a specific ESP syntax, rendering platform, or test suite.
- Saving results requires user confirmation and a host with write access to the memory path.
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 |
|---|---|---|---|---|
| Dynamic Email Personalizer this page | 60 · Recommended | ★ 3k | 13d ago | Apache-2.0 |
| Email Preference & Frequency Manager | 61 · Recommended | ★ 3k | 13d ago | Apache-2.0 |
| Reactivation Specialist | 57 · Use with care | ★ 3k | 13d ago | Apache-2.0 |
| Email List Segment Builder | 61 · Recommended | ★ 3k | 13d ago | Apache-2.0 |
| Email Lifecycle Flow Designer | 57 · Use with care | ★ 3k | 13d ago | Apache-2.0 |
How did FollowSkills review this skill?
The skill treats CSVs, ESP reports, and pasted rows as untrusted input, forbids following embedded instructions or echoing raw PII, and defines a PII blocklist, user confirmation before saving, optional connectors, and no autonomous sending. Deducted 7 points because the memory write/promotion path is complex, while rollback, complete permission isolation, and recovery after save failure are not explicit in this skill; the publisher is unverified, but that is not deducted by itself.
Input confirmation, NEEDS_INPUT handling for missing columns, fill-rate decision gates, mandatory fallbacks, mandatory catch-all branches, and abnormal-input guidance are fairly detailed. Deducted 11 points because the skill has no dedicated tests, scripts, or reproducible execution evidence; referenced files and host behavior are not verified here, and failure reporting and recovery remain partly delegated to the host.
Triggers, audience scenarios, inputs, four-part output, operating modes, and non-fit boundaries are clear, covering B2C, B2B cold outbound, and newsletters; Chinese users are partly supported and connectors are optional. Deducted 4 points because the core instructions are mainly English with limited Chinese-output constraints, while ESP syntax and cross-host compatibility require later handling; geo-relevance is declared low.
The skill has structured frontmatter, version 18.0.0, Apache-2.0 licensing, Quick Starts, a contract, data-source guidance, instructions, save-result guidance, and next-skill routing. Deducted 5 points because this skill does not clearly state changelog ownership, maintenance responsibility, or an update path, and several referenced files were not supplied for verification.
The intended deliverable is concrete and directly handoff-oriented: merge-tag mapping, rendered fallbacks, conditional rules, safety audit, and PII guard, with actionable completion criteria. Static calibration limits this to 7 because no real input, representative output, or execution result is provided to verify completeness, platform-syntax correctness, or marginal benefit over manual work.
The skill supplies explicit checks, done criteria, and Measured/Estimated labeling, and points to shared security and contract documents. Capped at 5 under static calibration: the supplied CI tests cover general architecture and audit-artifact validation rather than this skill's key paths, with no third-party execution evidence or independent corroboration.
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 →