agent-v3-performance-engineer

Agent skill for v3-performance-engineer - invoke with $agent-v3-performance-engineer

3.20x

Quality

Does it follow best practices?

Impact

96%

3.20x

Average score across 3 eval scenarios

Securityby

Passed

No known issues

Optimize this skill with Tessl

npx tessl skill review --optimize ./.agents/skills/agent-v3-performance-engineer/SKILL.md

Evaluation results

90%

71%

Benchmark Suite: Attention Mechanism & Vector Search

Attention and search benchmark targets

Criteria

Without context

With context

Attention speedup target range

100%

Attention memory reduction target

100%

Attention sequence lengths

100%

Attention result structure

62%

100%

Search improvement target range

100%

Search achieved flag

37%

100%

Search test query count

100%

Search result structure

50%

75%

Memory usage tracking

50%

100%

Memory reduction target

100%

Memory result structure

50%

66%

Sub-100ms latency note

100%

82%

Agent CLI Startup & Adaptive Learning Benchmarks

Startup and SONA adaptation benchmarks

Criteria

Without context

With context

Startup target 500ms

62%

100%

Startup return fields

40%

100%

SONA target 0.05ms

100%

hrtime.bigint() for SONA

100%

SONA scenario: pattern_recognition

100%

SONA scenario: task_optimization

100%

SONA scenario: error_correction

100%

SONA scenario: performance_tuning

100%

SONA scenario: behavior_adaptation

100%

agentic-flow store-pattern command

100%

store-pattern flags present

100%

SONA result structure

16%

100%

SONA successRate field

100%

47%

Performance Monitoring Dashboard & Regression Detection System

Regression detection and performance monitoring

Criteria

Without context

With context

MetricCollector key: flash_attention_speedup

100%

MetricCollector key: search_improvement

100%

MetricCollector key: memory_reduction

100%

MetricCollector key: startup_time

100%

MetricCollector key: sona_adaptation

100%

5% regression threshold

50%

100%

Regression return: hasRegressions

100%

Regression return: regressions array

100%

Regression return: recommendations

100%

PerformanceReport: summary field

100%

PerformanceReport: achievements field

100%

PerformanceReport: recommendations field

100%

PerformanceReport: trends field

100%

PerformanceReport: nextActions field

100%

15-agent swarm benchmark

100%

Swarm: 3 timing measurements

100%

Repository: ruvnet/claude-flow
Commit: f547cec

Evaluated: 26 days ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

Benchmark Suite: Attention Mechanism & Vector Search Agent CLI Startup & Adaptive Learning Benchmarks Performance Monitoring Dashboard & Regression Detection System

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.