performance-benchmarker

Runs load tests, profiles application bottlenecks, analyzes response time metrics, and generates performance optimization recommendations. Use when you need to benchmark a system, run load or stress tests, measure latency or throughput, identify performance bottlenecks, optimize Core Web Vitals (LCP, FID, CLS), plan capacity, enforce performance budgets in CI/CD pipelines, or investigate slow API response times, database query performance, or frontend rendering delays.

1.08x

Quality

88%

Does it follow best practices?

Impact

92%

1.08x

Average score across 3 eval scenarios

Securityby

Risky

Do not use without reviewing

Evaluation results

88%

20%

Load Test Script for ShopStream API

k6 load test script structure

Criteria

Without context

With context

Uses k6 tool

100%

Stage: warm-up

100%

Stage: peak and stress

100%

Stage: cool-down

100%

p95 threshold value

100%

Error rate threshold value

100%

handleSummary output file

BASE_URL env var

100%

sleep for think time

100%

k6 run command with JSON output

100%

checks on responses

100%

98%

API Performance Investigation for Payvault

Performance report and optimization workflow

Criteria

Without context

With context

SLA fail determination

100%

p50/p95/p99 in report

100%

Error rate in report

100%

Throughput in report

100%

Database bottleneck identified

100%

EXPLAIN ANALYZE recommendation

100%

One change at a time

100%

Statistical significance: 3 iterations

100%

Variance threshold

100%

Test configuration section

100%

Next steps section

100%

Reproduce before fixing

75%

91%

Performance Quality Gates for Luminos Media Platform

CI/CD performance gates and Core Web Vitals

Criteria

Without context

With context

LCP: preload hero image

100%

LCP: WebP or AVIF format

CLS: reserved space for images

100%

CLS: no above-fold injection

100%

FID/INP: break up long task

100%

77%

FID/INP: defer non-critical JS

100%

Measurement tools mentioned

100%

CI p95 threshold

55%

100%

CI error rate threshold

100%

CI build fails on threshold breach

100%

CI runs on pull requests

100%

k6 JSON output flag

100%

Repository: OpenRoster-ai/awesome-agents
Commit: 010799b

Evaluated: 20 days ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

Load Test Script for ShopStream API API Performance Investigation for Payvault Performance Quality Gates for Luminos Media Platform

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.