Ai marketing evaluator

Skill duandigi/duandigi-growth-marketing-skill/skills/ai-marketing-evaluator

Evidence-first AI Agent Skills for growth marketing — multi-channel analytics (SEO, paid media, social, local search, CRM), secure account integration, AI evaluation, experimentation & approval-safe optimization for Claude Code.

Install

npx -y skills add duandigi/duandigi-growth-marketing-skill --skill ai-marketing-evaluator

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

25 days oldThe repository was created 25 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Use this skill when producing an evidence-backed AI evaluation of a project, portfolio, channel, campaign, landing page, content group, funnel, lead source, or revenue outcome from normalized multi-source data.

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

3.6 KB, as published. Nobody here has run it

Ai Marketing Evaluator

Purpose

Explain what changed, why it may have changed, what the business impact is, what remains unknown, and which action or experiment should happen next.

Inputs

Project goals, growth model, and decision context
Connection health and data-quality assessment
Normalized channel, conversion, lead, and revenue data
Historical baseline, seasonality, changes, experiments, and constraints

If a required input is unavailable, label it unknown, state how it limits the decision, and create a collection, mapping, validation, or instrumentation task. Never invent credentials, assets, metrics, permissions, or business outcomes.

Workflow

Confirm the evaluation period, comparison period, data completeness, and decision owner.
Describe material changes from portfolio to project, channel, object, funnel stage, and business outcome.
Separate observed facts, calculated facts, plausible explanations, and unknowns.
Test measurement, mix, seasonality, campaign, landing-page, operational, and market explanations before selecting a likely cause.
Assess impact on customer value, qualified outcomes, revenue, efficiency, and risk rather than vanity metrics alone.
Create no more than five prioritized findings, each with evidence, confidence, missing data, and next action.
Convert eligible findings into alerts, research tasks, experiments, or approval requests.

Required output

Return a concise, decision-oriented result containing:

Executive evaluation and scope
Prioritized evaluation cards
Evidence, confidence, competing explanations, and missing data
Business impact and affected funnel stage
Recommended task, experiment, or decision with approval level

Label material statements as confirmed, calculated, inferred, assumed, or unknown. Include the data period, last complete period, source lineage, and confidence whenever they can change the decision.

Guardrails

Do not produce an AI score or causal explanation when connection health or data quality is insufficient.
Do not state that correlation, attribution, or before-after change proves causality.
Do not recommend optimization toward unqualified leads, platform conversions, or engagement if business outcomes worsen.
Do not hide uncertainty behind polished narrative.
Do not claim guaranteed growth or present an estimate as observed fact.
Do not reveal secrets, personal data, private provider payloads, or cross-project information.
When an action can spend money, publish, contact people, alter access, modify production, or delete data, prepare an approval request instead of executing automatically.

Completion check

Before finishing, verify that the output:

answers a specific business or implementation decision;
uses the correct organization, project, asset, date range, time zone, and currency;
separates performance problems from data, connection, and attribution problems;
includes evidence, uncertainty, affected scope, and a measurable next step;
respects least privilege, approval, audit, and rollback requirements;
is no longer than necessary for the decision.