From magic-powers
Synthesize an Amplitude dashboard into executive narrative with key findings, trends, and risks. Uses mcp__Amplitude__get_dashboard, mcp__Amplitude__query_charts.
npx claudepluginhub kienbui1995/magic-powers --plugin magic-powersThis skill uses the workspace's default tool permissions.
- A stakeholder asks for a summary of what a dashboard is showing
Generates design tokens/docs from CSS/Tailwind/styled-components codebases, audits visual consistency across 10 dimensions, detects AI slop in UI.
Records polished WebM UI demo videos of web apps using Playwright with cursor overlay, natural pacing, and three-phase scripting. Activates for demo, walkthrough, screen recording, or tutorial requests.
Delivers idiomatic Kotlin patterns for null safety, immutability, sealed classes, coroutines, Flows, extensions, DSL builders, and Gradle DSL. Use when writing, reviewing, refactoring, or designing Kotlin code.
Call mcp__Amplitude__get_dashboard to retrieve the dashboard configuration and chart list. Then use mcp__Amplitude__query_charts to fetch current data for all charts. Do not analyze charts individually in isolation — the goal is synthesis across the entire dashboard.
Note: query all charts before drawing conclusions. A metric that looks alarming in isolation may be explained by another chart on the same dashboard.
Look for the narrative thread connecting the charts:
The story is not a list of what each chart shows. It is the meaning that emerges when the charts are read together.
Select the most important findings across the entire dashboard. Rank by business impact:
For each finding, provide: the specific metric, the direction and magnitude of change, the time window, and why it matters to the business.
Flag any metric that deviates unexpectedly from its trend. For each anomaly:
If the dashboard includes qualitative signals (NPS, CSAT, support ticket volume), correlate them with the quantitative metrics:
Conclude with 2-4 concrete recommended actions. Each recommendation should:
mcp__Amplitude__get_dashboard — load dashboard structure and chart listmcp__Amplitude__query_charts — fetch current data for all charts in the dashboardmcp__Amplitude__render_chart — visual rendering of individual charts for pattern recognitionmcp__Amplitude__get_feedback_insights — correlate quantitative signals with user feedback (if available)Output is written in narrative paragraphs readable by executives — not labeled database fields or bullet dumps.
Structure:
Every factual claim must include a specific number. No vague language ("metrics look good," "some improvement").