--- artifact: skill-implementation-packet version: "1.0" repo_version: "2.10.0" skill_version: "1.0.0" created: 2026-04-11 status: sample thread: storevine context: Storevine B2B ecommerce platform . using the skill builder to create a campaign-analytics skill --- ## Scenario A contributor runs the skill builder to create a "campaign-analytics" skill for defining email campaign measurement frameworks. The builder checks overlap with existing measure skills, confirms the gap (no skill covers funnel metrics, cohort analysis, or attribution models), and generates a Skill Implementation Packet. [fictional] --- ## Prompt ``` /pm-skill-builder Idea: A skill for defining email campaign measurement frameworks -- funnel metrics (sent > delivered > opened > clicked > converted), cohort comparison, and revenue attribution. Existing measure skills handle dashboards and event tracking but not campaign-specific analytics frameworks. ``` --- ## Output # Skill Implementation Packet: campaign-analytics > **Created**: 2026-04-11 > **Builder version**: 1.0.0 > **Status**: Draft ## Decision Create new skill. Why Gate evidence: (1) `/dashboard-requirements` produced a dashboard spec but skipped funnel definition and attribution; (2) `/instrumentation-spec` produced event schemas but not the analytical framework connecting events to outcomes; (3) `/experiment-design` assumed A/B test structure rather than ongoing measurement. [fictional] ## Classification | Field | Value | |-------|-------| | Type | domain | | Phase | measure | | Directory | measure-campaign-analytics | | Command | /campaign-analytics | ## Overlap Analysis ### Existing skills checked measure-dashboard-requirements, measure-instrumentation-spec, measure-experiment-design, measure-experiment-results, deliver-acceptance-criteria, define-hypothesis ### Findings Existing skills assume the PM already knows which metrics matter or are scoped to A/B tests. None address funnel definition, cohort segmentation, or revenue attribution. [fictional] ### Why this skill is still needed Fills the gap between "we need to measure campaigns" and "here is the dashboard spec" -- produces the analytical foundation that feeds downstream skills. [fictional] ## Quality Forecast | Zone | Weight | Description | |------|--------|-------------| | Knowledge (K) | 35 | Funnel metrics, attribution models, cohort analysis | | Process (P) | 40 | Step-by-step framework definition | | Constraint (C) | 15 | Vanity metric guardrails and attribution pitfalls | | Wisdom (W) | 10 | Attribution model selection judgment | **Dominant zone**: Process (40%). **Exemplars**: `measure-dashboard-requirements`, `measure-instrumentation-spec`. ## Draft Frontmatter ```yaml --- name: campaign-analytics description: Defines a measurement framework for email campaign performance covering funnel metrics, cohort analysis, KPI specification, and revenue attribution. phase: measure version: "1.0.0" updated: 2026-04-11 license: Apache-2.0 metadata: category: specification --- ``` ## Draft Files (abbreviated) **SKILL.md**: Guides PMs through defining funnel stages, KPIs, cohort comparisons, and attribution models. 6 instruction steps, output contract, quality checklist. [fictional] **TEMPLATE.md**: 7 sections -- Campaign Context, Funnel Definition, KPI Specification, Cohort Design, Attribution Model, Data Source Mapping, Open Questions. [fictional] **EXAMPLE.md**: Worked example (180 lines) for a B2B SaaS onboarding email sequence. [fictional] **Command + AGENTS.md**: Standard routing to `skills/measure-campaign-analytics/SKILL.md`. [fictional] ## Validation Checklist All 9 CI checks pass (name, description, version, updated, license, phase consistency, TEMPLATE.md sections, command path, AGENTS.md format). All 6 quality checks pass (EXAMPLE.md 180 lines, output contract, quality checklist, dominant zone identified, gap analysis on 6 skills, Why Gate names 3 prompts). [fictional] ## Next Steps 1. Review the packet and flag changes needed. 2. Run `bash scripts/lint-skills-frontmatter.sh` after promotion. 3. Test `/campaign-analytics` with a realistic scenario, then use `/pm-skill-iterate` to refine.