home / skills / levnikolaevich / claude-code-skills / ln-774-healthcheck-setup
/ln-774-healthcheck-setup
This skill configures Kubernetes health probes for readiness, liveness, and startup, auto-detecting dependencies and generating endpoints for resilient apps.
npx playbooks add skill levnikolaevich/claude-code-skills --skill ln-774-healthcheck-setupReview the files below or copy the command above to add this skill to your agents.
---
name: ln-774-healthcheck-setup
description: Configures health check endpoints for Kubernetes readiness/liveness/startup
---
# ln-774-healthcheck-setup
**Type:** L3 Worker
**Category:** 7XX Project Bootstrap
**Parent:** ln-770-crosscutting-setup
Configures health check endpoints for Kubernetes probes and monitoring.
---
## Overview
| Aspect | Details |
|--------|---------|
| **Input** | Context Store from ln-770 |
| **Output** | Health check endpoints and Kubernetes probe configuration |
| **Stacks** | .NET (AspNetCore.Diagnostics.HealthChecks), Python (FastAPI routes) |
---
## Phase 1: Receive Context + Identify Dependencies
Accept Context Store and scan for dependencies to monitor.
**Required Context:**
- `STACK`: .NET or Python
- `PROJECT_ROOT`: Project directory path
**Idempotency Check:**
- .NET: Grep for `AddHealthChecks` or `MapHealthChecks`
- Python: Grep for `/health` route
- If found: Return `{ "status": "skipped" }`
**Dependency Detection:**
| Dependency | .NET Detection | Python Detection |
|------------|----------------|------------------|
| PostgreSQL | `Npgsql` in csproj | `psycopg2` or `asyncpg` in requirements |
| MySQL | `MySql.Data` in csproj | `mysql-connector-python` in requirements |
| Redis | `StackExchange.Redis` in csproj | `redis` in requirements |
| RabbitMQ | `RabbitMQ.Client` in csproj | `pika` or `aio-pika` in requirements |
| MongoDB | `MongoDB.Driver` in csproj | `pymongo` in requirements |
---
## Phase 2: Design Health Check Strategy
Define three types of health endpoints per Kubernetes best practices.
### Endpoint Types
| Endpoint | Probe Type | Purpose | Checks |
|----------|------------|---------|--------|
| `/health/live` | Liveness | Is app alive? | App responds (no dependency checks) |
| `/health/ready` | Readiness | Can app serve traffic? | All dependencies healthy |
| `/health/startup` | Startup (K8s 1.16+) | Is app initialized? | Initial warmup complete |
### When Each Probe Fails
| Probe | Failure Action | Kubernetes Behavior |
|-------|----------------|---------------------|
| Liveness | Container restart | kubelet restarts container |
| Readiness | Remove from service | Traffic stopped, no restart |
| Startup | Delay other probes | Liveness/Readiness paused |
---
## Phase 3: Research Health Check Patterns
Use MCP tools for current documentation.
**For .NET:**
```
MCP ref: "ASP.NET Core health checks Kubernetes probes"
Context7: /dotnet/aspnetcore
```
**For Python:**
```
MCP ref: "FastAPI health check endpoint Kubernetes"
Context7: /tiangolo/fastapi
```
**Key Patterns to Research:**
1. Database health checks (connection pool)
2. Redis connectivity check
3. Custom health check implementation
4. Health check response writer customization
---
## Phase 4: Configure Kubernetes Probes
Determine probe timing based on application characteristics.
### Probe Configuration
| Parameter | Liveness | Readiness | Startup |
|-----------|----------|-----------|---------|
| `initialDelaySeconds` | 10 | 5 | 0 |
| `periodSeconds` | 10 | 5 | 5 |
| `timeoutSeconds` | 5 | 3 | 3 |
| `failureThreshold` | 3 | 3 | 30 |
| `successThreshold` | 1 | 1 | 1 |
**Startup Probe Calculation:**
```
Max startup time = initialDelaySeconds + (periodSeconds × failureThreshold)
Default: 0 + (5 × 30) = 150 seconds
```
---
## Phase 5: Generate Implementation
### .NET Output Files
| File | Purpose |
|------|---------|
| `Extensions/HealthCheckExtensions.cs` | Health check registration |
| `HealthChecks/StartupHealthCheck.cs` | Custom startup check |
**Generation Process:**
1. Use MCP ref for current ASP.NET Core health checks API
2. Generate HealthCheckExtensions with:
- AddHealthChecks registration
- Database health check (if detected)
- Redis health check (if detected)
- Custom StartupHealthCheck
3. Configure three endpoints with proper tags
**Packages to Add:**
- `AspNetCore.HealthChecks.NpgSql` (if PostgreSQL)
- `AspNetCore.HealthChecks.Redis` (if Redis)
- `AspNetCore.HealthChecks.MySql` (if MySQL)
**Registration Code:**
```csharp
builder.Services.AddHealthCheckServices(builder.Configuration);
// ...
app.MapHealthCheckEndpoints();
```
### Python Output Files
| File | Purpose |
|------|---------|
| `routes/health.py` | Health check router |
| `services/health_checker.py` | Dependency health checks |
**Generation Process:**
1. Use MCP ref for FastAPI health patterns
2. Generate health router with:
- /health/live endpoint (simple)
- /health/ready endpoint (with dependency checks)
- /health/startup endpoint
3. Generate health_checker service for dependency verification
**Registration Code:**
```python
from routes.health import health_router
app.include_router(health_router)
```
### Kubernetes Manifest Snippet
Generate for inclusion in deployment.yaml:
```yaml
livenessProbe:
httpGet:
path: /health/live
port: 5000
initialDelaySeconds: 10
periodSeconds: 10
timeoutSeconds: 5
failureThreshold: 3
readinessProbe:
httpGet:
path: /health/ready
port: 5000
initialDelaySeconds: 5
periodSeconds: 5
timeoutSeconds: 3
failureThreshold: 3
startupProbe:
httpGet:
path: /health/startup
port: 5000
periodSeconds: 5
failureThreshold: 30
```
---
## Phase 6: Validate
**Validation Steps:**
1. **Syntax check:**
- .NET: `dotnet build --no-restore`
- Python: `python -m py_compile routes/health.py`
2. **Endpoint test:**
```bash
curl http://localhost:5000/health/live
curl http://localhost:5000/health/ready
curl http://localhost:5000/health/startup
```
3. **Verify response format:**
```json
{
"status": "Healthy",
"checks": {
"database": { "status": "Healthy", "duration": "00:00:00.0234" },
"redis": { "status": "Healthy", "duration": "00:00:00.0012" }
},
"totalDuration": "00:00:00.0250"
}
```
4. **Dependency failure test:**
- Stop database
- Verify `/health/ready` returns 503
- Verify `/health/live` still returns 200
---
## Return to Coordinator
```json
{
"status": "success",
"files_created": [
"Extensions/HealthCheckExtensions.cs",
"HealthChecks/StartupHealthCheck.cs"
],
"packages_added": [
"AspNetCore.HealthChecks.NpgSql"
],
"registration_code": "builder.Services.AddHealthCheckServices(configuration);",
"message": "Configured health checks with liveness, readiness, and startup probes"
}
```
---
## Reference Links
- [ASP.NET Core Health Checks](https://learn.microsoft.com/aspnet/core/host-and-deploy/health-checks)
- [Kubernetes Probes](https://kubernetes.io/docs/tasks/configure-pod-container/configure-liveness-readiness-startup-probes/)
- [AWS EKS Health Check Best Practices](https://docs.aws.amazon.com/eks/latest/best-practices/application.html)
---
## Critical Rules
- **Three separate endpoints** — `/health/live`, `/health/ready`, `/health/startup` per Kubernetes best practices
- **Liveness must not check dependencies** — only confirms app is alive (avoids cascade restarts)
- **Readiness checks all dependencies** — DB, Redis, RabbitMQ connectivity verified
- **Auto-detect dependencies from project files** — scan csproj/requirements for known packages
- **Idempotent** — if `AddHealthChecks`/`MapHealthChecks` or `/health` route exists, return `status: "skipped"`
## Definition of Done
- Context Store received (stack, project root)
- Dependencies detected (PostgreSQL, MySQL, Redis, RabbitMQ, MongoDB)
- Health check endpoints generated (live, ready, startup) for detected stack
- Kubernetes probe manifest snippet generated with proper timing parameters
- Syntax validated (`dotnet build` or `py_compile`)
- Structured JSON response returned to ln-770 coordinator
---
**Version:** 2.0.0
**Last Updated:** 2026-01-10
This skill configures health check endpoints and Kubernetes probe settings for applications using .NET or Python. It auto-detects common dependencies, generates liveness/readiness/startup endpoints, and emits a recommended probe manifest and registration code. The goal is a production-ready, idempotent setup that follows Kubernetes best practices for probes.
The skill accepts a Context Store with STACK and PROJECT_ROOT, scans project files to detect dependencies (Postgres, MySQL, Redis, RabbitMQ, MongoDB), and checks for existing health endpoints to remain idempotent. It selects patterns for the target stack (.NET or FastAPI), generates registration and router files, adds dependency-specific health checks, and produces a Kubernetes probe snippet. Finally it runs simple validations: build/compile checks and curl tests for the three endpoints.
How does the skill detect dependencies?
It scans csproj files for known package names for .NET and requirements files for Python package names to identify databases, caches, and brokers.
What if health checks already exist?
The skill performs an idempotency check; if AddHealthChecks/MapHealthChecks or a /health route is found it returns skipped to avoid duplicates.