send-experiment-designer

Use when the user asks to "design an email A/B test", "set up a multivariate subject/CTA test", "run a send-time test", "build a hold-out group", or "is this email result statistically and practically material?"; produces a falsifiable hypothesis, one-variable-per-cell matrix, sample-size/MDE/duration/power plan, and an effect/uncertainty read from own ESP data. Applies only a precommitted owner-approved action rule; the helper never chooses a business action. Not for EQS/vetoes or writing the email. 邮件AB测试设计/多变量测试/发送时间测试/留出组/显著性判定

Quality

92%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

85%Weight 40%Scale 1-3

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A well-structured, actionable skill body with a strong stepwise workflow, validation checkpoints, and clean one-level-deep references. Its only real weakness is some repetition between the mode table and step 3 that a tightening pass would remove.

Suggestions

De-duplicate the mode descriptions: keep the Quick Start table as the canonical reference and have step 3 link back to it instead of restating each mode.

Consolidate the repeated 'keep a control in every design' / control-sizing guidance into a single stated rule rather than restating it across steps 3, 5, and 6.

Trim the duplicated scope-guard text between the intro paragraph and the Skill Contract section to recover tokens without losing the boundary.

Dimension	Reasoning	Score
Conciseness	Dense and substantive but some content repeats (the mode table re-described in step 3, 'keep a control in every design' stated several times) and could be tightened; it is mostly efficient but carries redundancy.	2 / 3
Actionability	Provides a copy-pasteable command (`python3 "${CLAUDE_PLUGIN_ROOT}/scripts/connectors/experiment.py" proportion ...`), named subcommands (`continuous`, `samplesize`), an explicit duration formula, and a concrete sample-size reference table — fully executable guidance.	3 / 3
Workflow Clarity	A 9-step sequence with explicit checkpoints and feedback loops (NEEDS_INPUT stop in step 1, underpowered-collapse fallback in step 6, multiple-comparison adjustment in step 7, UNDECIDED decision gate with guardrails in step 8).	3 / 3
Progressive Disclosure	The body is an overview pointing one level deep to clearly signaled references (send-benchmark.md, measurement-protocol.md, skill-contract.md, CONNECTORS.md, SECURITY.md, experiment.py) with inline guidance kept appropriately concise.	3 / 3
	Total	11 / 12 Passed

Description

100%Weight 40%Scale 1-3

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A high-quality description: third-person, concrete outputs, natural user triggers, and an explicit scope boundary distinguishing it from sibling skills. Nothing material is missing or vague.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete outputs ('falsifiable hypothesis, one-variable-per-cell matrix, sample-size/MDE/duration/power plan, and an effect/uncertainty read'), matching the anchor for listing several specific concrete actions.	3 / 3
Completeness	Clearly answers both what (the produced design/read-out artifacts) and when ('Use when the user asks to...'), and adds an explicit negative scope ('Not for EQS/vetoes or writing the email').	3 / 3
Trigger Term Quality	Natural trigger phrasings a user would actually say ('design an email A/B test', 'set up a multivariate subject/CTA test', 'run a send-time test', 'build a hold-out group') give good coverage of common variations.	3 / 3
Distinctiveness Conflict Risk	A clear email-experiment niche with distinct triggers plus an explicit out-of-scope boundary makes it unlikely to fire for the wrong skill.	3 / 3
	Total	12 / 12 Passed

Validation

81%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_field	'metadata' should map string keys to string values	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 19 suspicious	Warning

	Total	13 / 16 Passed

Repository: aaron-he-zhu/aaron-marketing-skills
Path: email/deliver/send-experiment-designer/SKILL.md
Commit: 4a77c48

Reviewed: 9 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.