benchmark

Run Performance Benchmarks

Usage: /benchmark or /benchmark --quick

Purpose: Execute performance benchmarks and analyze results

---

🎯 What This Command Does

Runs the project's performance benchmark suite using the existing Makefile command:

make benchmark

This executes benchmarks for:

• Agent response time
• LLM call latency
• Authorization check performance
• Database query performance
• Context loading speed

---

📊 Execution Steps

Step 1: Run Benchmarks

Execute the benchmark suite:

# Full benchmark suite
make benchmark

# View benchmark configuration
cat tests/benchmark/config.yaml

Step 2: Parse Results

Analyze benchmark output:

# Benchmark results are typically in:
# - pytest output (console)
# - .benchmark/ directory (if configured)

# Look for metrics like:
# - ops/sec (operations per second)
# - mean, median, p95, p99 latencies
# - memory usage
# - CPU usage

Step 3: Compare with Baselines

# If baseline metrics exist
if [ -f "tests/regression/baseline_metrics.json" ]; then
    echo "Comparing with baseline..."
    # Check for regressions (>20% slower)
fi

Step 4: Generate Summary

Provide user with formatted summary:

=== Performance Benchmark Results ===

Benchmark Suite: Agent Performance
Executed: YYYY-MM-DD HH:MM:SS

Results:
┌─────────────────────────────┬──────────┬─────────┬─────────┐
│ Benchmark                   │ Mean     │ P95     │ Status  │
├─────────────────────────────┼──────────┼─────────┼─────────┤
│ agent_response              │ 2.3s     │ 4.5s    │ ✅ PASS │
│ llm_call                    │ 1.8s     │ 3.2s    │ ✅ PASS │
│ authorization_check         │ 25ms     │ 45ms    │ ✅ PASS │
│ database_query              │ 15ms     │ 30ms    │ ✅ PASS │
│ context_loading             │ 120ms    │ 200ms   │ ✅ PASS │
└─────────────────────────────┴──────────┴─────────┴─────────┘

Baseline Comparison:
- No regressions detected ✅
- 2/5 benchmarks improved (context_loading: -15ms)

Memory Usage:
- Peak: 256 MB
- Average: 180 MB

Overall Status: ✅ ALL BENCHMARKS PASSING