ElBruno.MempalaceNet: Skill for Claude Code

Skill: Performance Tracking with MemPalace.Diagnostics

Status: Active
Domain: Diagnostics, SLA Validation, Benchmarking
Created: 2026-04-28
Author: Rachael (CLI/UX Dev)

---

Overview

Use MemPalace.Diagnostics.PerformanceBenchmark to track operation latencies, calculate percentile statistics (P50/P95/P99/P100), validate SLA compliance, and generate human/machine-readable reports.

---

When to Use This Skill

• ✅ Tracking latency of API calls, database queries, AI inference, search operations
• ✅ Validating P95/P99 SLA compliance in integration tests or CI pipelines
• ✅ Generating performance reports for trend analysis or debugging
• ✅ Comparing performance across different implementations (e.g., A/B testing)
• ✅ Monitoring long-running services with periodic latency snapshots

❌ Do NOT use for:

• Real-time metrics collection in production (use OpenTelemetry, Prometheus instead)
• Memory profiling or CPU profiling (use dotnet-trace, PerfView instead)
• Distributed tracing (use Application Insights, Jaeger instead)

---

Pattern 1: Basic Latency Tracking

Use Case: Measure and validate a single operation's P95 latency.

using MemPalace.Diagnostics;
using System.Diagnostics;

[Fact]
public async Task SearchOperation_MeetsP95SLA()
{
    var benchmark = new PerformanceBenchmark();
    
    // Run 100 search operations
    for (int i = 0; i < 100; i++)
    {
        var sw = Stopwatch.StartNew();
        await _searchService.Search("test query");
        benchmark.RecordLatency("search", sw.Elapsed);
    }
    
    // Validate P95 < 100ms
    var passes = benchmark.ValidateSLA("search", TimeSpan.FromMilliseconds(100));
    
    if (!passes)
    {
        var stats = benchmark.GetPercentiles("search");
        Assert.Fail($"Search P95 SLA failed: {stats.P95.TotalMilliseconds:F2}ms exceeds 100ms");
    }
}

---

Pattern 2: Multi-Operation Batch Validation

Use Case: Validate multiple operations against their SLAs in a single test.