From magic-powers
B2B account health assessment covering usage patterns, expansion risk, and growth opportunities. Uses mcp__Amplitude__get_users, mcp__Amplitude__query_amplitude_data.
npx claudepluginhub kienbui1995/magic-powers --plugin magic-powersThis skill uses the workspace's default tool permissions.
- A customer success manager needs a data-driven health check before a QBR or renewal conversation
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.
Find the account using mcp__Amplitude__get_users with identifiers provided by the requester:
Retrieve the full list of user profiles associated with this account. Note the total user count, when accounts were created, and any user properties that indicate role or plan tier.
If user profiles are sparse or the account is hard to identify, ask the requester for the organization ID or a specific admin email as an anchor.
Use mcp__Amplitude__query_amplitude_data to compute the core health signals. For each signal, assess: Green (healthy), Yellow (monitor), or Red (at risk).
Health signal thresholds (adjust to your product's benchmarks):
| Signal | Green | Yellow | Red |
|---|---|---|---|
| DAU/MAU ratio | >0.20 | 0.10-0.20 | <0.10 |
| Weekly active users / total licensed users | >60% | 30-60% | <30% |
| Feature breadth (features used in last 30d) | Top 50% of accounts | Middle 25% | Bottom 25% |
| Activity trend (last 30d vs prior 30d) | Up or stable | -10% to 0% | <-10% |
| Days since last active session (power users) | <3 | 3-7 | >7 |
Also check: any error events, export failures, or API error patterns that suggest technical friction.
Go below account-level to understand who is active and who is dormant:
For each champion, note: their activity level, which features they use most, and whether their usage is growing or declining.
Understand which features drive stickiness vs which are underutilized:
Sticky features: features used frequently by champions. High usage of sticky features correlates with retention and expansion.
Underutilized features: features the account has access to but rarely uses. These represent:
Distinguish between the two by looking at whether similar accounts in the same segment heavily use these features. If yes, this account has an adoption gap worth addressing.
Usage depth indicators:
Look for signals that suggest the account is ready for expansion:
For each expansion signal, estimate the expansion opportunity: what plan upgrade or add-on is most appropriate?
mcp__Amplitude__get_users — find user profiles associated with the accountmcp__Amplitude__query_amplitude_data — compute usage metrics, DAU/MAU, feature adoptionmcp__Amplitude__get_session_replays — review sessions of churning or struggling users (optional)mcp__Amplitude__get_feedback_insights — surface any feedback from this account's usersmcp__Amplitude__get_context — get projectId and organization context (always first)The output is structured for a customer success or account management audience — practical and action-oriented.
Structure: