Email Preference & Frequency Manager
Design preference centers and opt-down paths that reduce fatigue before hard unsubscribe.
The skill treats exports as untrusted data and explicitly forbids following embedded instructions. It keeps the hard unsubscribe reachable in one click and separates responsibilities among this skill, consent-registry, and the auditor. Its read/write/promote flows and confirmation-gated memory save are disclosed, with no credential handling or direct outbound send. Deducted 5 points because runtime memory writes lack a fully specified permission boundary, rollback path, and sensitive-data treatment in this file; publisher identity is unverified.
The contract, inputs, outputs, done criteria, stop conditions, and missing-export behavior are clearly stated. Deducted 1 point because key behavior depends on unavailable companion skills, actual ESP configuration, and manual mappings; there are no skill-specific tests, schema checks, or diagnostic failure examples. Static calibration caps this dimension at 10, so it remains below that cap.
Trigger semantics, use cases, non-fit boundaries, parameters, and expected outputs are explicit. Tier-1 operation can use user-supplied data, and the core function does not entirely depend on unreachable overseas services; Chinese labels and bilingual hints are present. Deducted 4 points because the body is predominantly English, platform/legal boundaries remain user-dependent, and real ESP capability differences are not fully specified.
The file has front matter, versioning, compatibility, Quick Starts, a shared contract, data sources, instructions, scope guards, next-skill routing, and a save-results reference, plus Apache-2.0 licensing. Deducted 5 points because it lacks a skill-specific changelog, named maintenance owner, and explicit update path; SECURITY.md says 17.x is the supported current line while this skill declares 18.0.0; many operational assumptions are delegated to external references.
The skill specifies a directly usable field set, opt-down ladder, preference-to-suppression/frequency mapping, and SEND N sub-item note, including minimum cadence tiers, pause, and reachable unsubscribe. Deducted 8 points because the files contain methodology and prompts but no representative completed outputs or third-party execution evidence; implementation still requires ESP, lifecycle, and consent-registry work. Static calibration caps this dimension at 7, so it receives 7.
Revision-pinned scope, explicit references, source-type labels, and required Measured/User-provided/Estimated distinctions provide some auditability. Repository CI and tests exist, but the supplied tests do not cover this skill's key paths. Deducted 6 points because there are no skill-specific reproducible examples, independent validation results, or third-party corroboration; this conclusion is based only on static documentation. Static calibration caps this dimension at 5, so it receives 4.
- The skill and ESP integration were not executed; actual field, frequency-cap, and suppression behavior is unverified.
- Runtime writes to memory depend on host hooks and shared protocols; deployment should verify permissions, privacy boundaries, failure recovery, and prevention of unintended canonical consent writes.
- Reconcile the 18.0.0 skill version with SECURITY.md stating that 17.x is the supported current line, and verify referenced companion skills and maintenance paths.
What does this skill do, and when should you use it?
This is the email-nurture skill for designing a subscriber-facing preference center and frequency opt-down ladder. It defines topic groups, cadence tiers, pause options, channel toggles, and the suppression or frequency rules associated with each choice. It also emits the SEND N sub-item note for preference-center and frequency options offered, but it does not compute EQS or rule the N1 unsubscribe veto. It fits teams with a defined email program that want a lighter-touch exit path before subscribers leave entirely.
Reads user-provided topic sets, cadence tiers, target segments, and optionally a manual ESP export; inventories the current unsubscribe path; defines topic groups, at least two cadence tiers, and a pause option; orders the opt-down ladder; maps every topic, cadence, and pause choice to an ESP and consent-registry suppression or frequency rule; defines the ladder's sunset terminus; and produces a preference-center field specification, opt-down ladder, choice-to-rule mapping, SEND N sub-item note, and handoff summary.
- An email team sees rising opt-out rates in a segment and needs frequency reduction or pause choices before hard unsubscribe.
- A brand sends product updates, promotions, and event invitations and wants subscribers to control those streams independently.
- A lifecycle marketer needs weekly, monthly, and pause selections translated into executable ESP send limits.
- A team is redesigning its unsubscribe page and wants to preserve one-click hard unsubscribe while offering clear step-down paths.
What are this skill's strengths and limitations?
- Produces a field specification, ordered opt-down ladder, and explicit rule mapping for implementation.
- Requires at least two cadence tiers, a pause option, and an always-reachable one-click hard unsubscribe.
- Runs at Tier 1 from user-provided inputs; ESP APIs are optional conveniences.
- Has clear boundaries and does not replace lifecycle-flow design or EQS auditing.
- Does not own or persist the final consent, unsubscribe, or suppression record; consent-registry does.
- Does not design lifecycle flow maps, global send caps, quiet hours, or cadence governance.
- The source does not document a vendor-specific field compatibility matrix or platform-level execution validation.
- Reliable topic design requires the user to provide or confirm the actual content streams.
How do you install this skill?
Install the repository bundle to use this skill. 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 it with git clone https://github.com/aaron-he-zhu/aaron-marketing-skills. The skill file is email/nurture/preference-frequency-manager/SKILL.md.
How do you use this skill?
Describe a concrete goal to the host, for example: Build a preference center for [product/audience] on [ESP]. Offer topics [list], cadence tiers [weekly/monthly], and a pause option instead of a hard unsubscribe. Include the audience, topic set, ESP, and existing cadence when available. If the topic set is genuinely unknowable and no export is available, the skill asks for topic and cadence choices; a missing ESP export does not block design and can be marked N/A for current-state findings.
How does this skill compare with similar options?
Unlike email-sequence-designer, this skill owns the preference center and per-subject opt-down rules, not lifecycle flows or global cadence governance. Unlike consent-registry, it designs the choice-to-rule mapping but does not own canonical consent or suppression records. Unlike email-quality-auditor, it emits only the SEND N sub-item note and does not compute EQS or rule N1.