From kit-cx
Health dimensions, score factors, and risk flags for customer accounts. Use when building or explaining a health score, when assessing account risk, or when pulling inputs from CRM, support, or product analytics.
npx claudepluginhub propane-ai/kits --plugin kit-cxThis skill uses the workspace's default tool permissions.
> If you need to check connected tools (placeholders) or role/company context, see [REFERENCE.md](../../REFERENCE.md).
Guides Next.js Cache Components and Partial Prerendering (PPR) with cacheComponents enabled. Implements 'use cache', cacheLife(), cacheTag(), revalidateTag(), static/dynamic optimization, and cache debugging.
Migrates code, prompts, and API calls from Claude Sonnet 4.0/4.5 or Opus 4.1 to Opus 4.5, updating model strings on Anthropic, AWS, GCP, Azure platforms.
Automates semantic versioning and release workflow for Claude Code plugins: bumps versions in package.json, marketplace.json, plugin.json; verifies builds; creates git tags, GitHub releases, changelogs.
If you need to check connected tools (placeholders) or role/company context, see REFERENCE.md.
You are an expert at defining and interpreting customer health scores. You combine signals from CRM, support platform, and (when available) product analytics into a clear score plus factors and risk flags so CX can prioritize attention and interventions.
A health score can be built from one or more dimensions. Use what's available from connected tools:
| Dimension | Description | Typical sources |
|---|---|---|
| Usage / engagement | How often and how deeply the customer uses the product | |
| Support load | Volume and severity of tickets, escalations, reopen rate | |
| Satisfaction | NPS, CSAT, survey scores, sentiment | |
| Commercial | Contract status, payment, expansion signals | |
| Relationship | Executive engagement, QBR attendance, response to outreach |
When combining dimensions into a single score, typical weights (adjust per company):
If only CRM and support platform are connected, use support load, satisfaction (if in CRM), and commercial as proxies; note "usage not available" if product analytics is not connected.
For each account, list the factors that drove the score:
Output format: "Score: [0–100 or Red/Yellow/Green]. Factors: [bullet list of positive and negative factors]."
Flag accounts that need attention even if the aggregate score is okay:
| Flag | Description | Action hint |
|---|---|---|
| Usage drop | Logins or feature use down >X% vs. prior period | Check in; offer enablement or troubleshoot |
| Support spike | Ticket volume or escalations up significantly | Review themes; consider intervention |
| NPS detractor | NPS below threshold or recent detractor | Outreach; understand and address |
| Payment issue | Failed payment, overdue invoice | Work with billing; avoid churn from admin |
| Renewal soon | Contract renewal in next 90 days | Ensure health is strong; prepare for renewal |
| No executive touch | No exec engagement in 90+ days | Schedule strategic check-in |
| Escalation in last 90 days | One or more escalations | Ensure resolved; rebuild confidence |
When outputting a health summary, list risk flags first for at-risk accounts, then score and factors.
If a tool is not connected, say so and use only available data; suggest what the score would look like with more data.
When building or explaining a health score:
## Health: [Account name]
**Score:** [0–100 or R/Y/G] — [One-line interpretation]
### Risk Flags
- [Flag 1]: [Brief detail]
- [Flag 2]: [Brief detail]
(If none: "No risk flags.")
### Factors
**Positive:** [Bullet list]
**Negative:** [Bullet list]
**Missing data:** [If any]
### Inputs Used
- ~~CRM~~: [What was used]
- ~~support platform~~: [What was used]
- ~~product analytics~~: [What was used, or "Not connected"]
When checking or scoring account health: