performance-optimization

A measurement-first guide to improving the performance of Spring Boot 4 and Java applications. It covers ways to collect evidence about CPU use, memory, database queries, response times, and throughput before changing code.

In plain words
What is it for?
Use it when investigating slow endpoints, database queries, memory or garbage-collection issues, connection-pool sizing, caching, or virtual-thread problems.
Why use it?
It helps distinguish a measured performance problem from a guess and gives each optimization a clear target and way to verify the result.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/loiane/specs-driven-development-spring-angular/performance-optimization
Any agent
npx skills add loiane/specs-driven-development-spring-angular --skill performance-optimization
Clone the repo
git clone --depth 1 https://github.com/loiane/specs-driven-development-spring-angular

Made for: Claude Code, Codex.

Per session 89 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,614 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00089 $0.02614
Opus 5 $0.00044 $0.01307
Sonnet 5 $0.00018 $0.00523
Haiku 4.5 $0.00009 $0.00261

Measured yesterday against content hash 06b7478c0468, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

performance-optimization scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/skills/performance-optimization/SKILL.md · 164 lines

How it starts

The opening of the file, as written. The whole thing — 164 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Performance optimization (Spring Boot 4 / JVM)

Rule zero: measure first

No optimization without a profile or a benchmark. "It feels slow" is a hypothesis, not a finding.

If a PR labeled perf lands without a profile artifact, a JMH result, a flame graph, or a before/after metric screenshot, request changes. The artifact path goes in the PR description and in 05-implementation-log.md.

Acceptable evidence:

  • async-profiler flame graph (CPU or alloc) saved as SVG/HTML.
  • JDK Flight Recorder recording (.jfr) with the relevant view captured.
  • JMH benchmark output (*.json from -rf json) with before/after.
  • Micrometer histogram screenshots (p50/p95/p99) from before and after.
  • Database EXPLAIN ANALYZE output for the query in question.

SLO-first, not optimization-first

Every perf claim is tied to a Service Level Objective. If there is no stated SLO, the work has no exit criterion — write the SLO first.

Template (the user fills the blanks; never invent the numbers):

Endpoint:        <method> <path>
p50 latency:     ___ ms   (current: ___ ms)
p95 latency:     ___ ms   (current: ___ ms)
p99 latency:     ___ ms   (current: ___ ms)
Throughput:      ___ rps  (current: ___ rps)
Error budget:    ___ %    (window: ___ )
Measurement:     <Micrometer metric name + dashboard link>

If the user has not committed to numbers, file a Q-NNN in 01-spec.md or 03-design.md and halt.

Tools

Tool Use for Notes
async-profiler CPU and allocation flame graphs in production-like envs -e cpu or -e alloc; safe in prod (sample-based)
JFR + JDK Mission Control Always-on low-overhead recording; method profiling, allocation, GC, locks Enable with -XX:StartFlightRecording=...
JMH Microbenchmarks for hot loops, parsers, serialization, comparator/equality changes Fork ≥ 2, warmup ≥ 5, measure ≥ 5; run on the deploy hardware class
Micrometer + Spring Boot Actuator Production metrics (latency histograms, throughput, error rate) Use Timer with publishPercentileHistogram(true); not Counter for latency
EXPLAIN ANALYZE Real query plan with row counts, buffers, timing PostgreSQL: EXPLAIN (ANALYZE, BUFFERS, VERBOSE) ...
-Djdk.tracePinnedThreads=full Detect virtual-thread pinning during dev/test Enable in test profiles when investigating throughput regressions
-Xlog:gc* or JFR GC events GC pause analysis Look at p99 pause, not average

Read the full file on GitHub · 164 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 164 lines · 89 tokens per session scan A 06b7478c0468

Subscribe to this mod's changes

performance-optimization is a skill published in the GitHub repository loiane/specs-driven-development-spring-angular (57 stars, last pushed 2mo ago), licensed MIT. It adds 89 tokens to every session and 2,614 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

obsidian-writer

Write well-formatted notes to the atmosphere-vault Obsidian knowledge base. Use this skill whenever creating or updating an ADR, runbook, plan, API doc, guide, session output, or any structured document that should land in the vault — even when the user doesn't say "Obsidian" explicitly. Delegates to…

Atmosphere/atmosphere · 93 tokens

llm-judge

AI quality judge that scores agent responses 0-10 across helpfulness, accuracy, completeness, and clarity. Use when evaluating multi-agent output or implementing LLM-as-judge quality gates.

Atmosphere/atmosphere · 43 tokens

spec-kitty-mission-review

Review a fully merged Spec Kitty mission post-merge (all WPs done/approved) to verify spec→code fidelity, FR coverage, drift, risks, and security. Triggers: "review the merged mission", "post-merge mission review", "verify the completed mission", "audit the mission implementation", "mission-level acceptance review"…

Priivacy-ai/spec-kitty · 155 tokens

spec-kitty-charter-doctrine

Run charter interview, generation, context, and sync workflows for project governance in Spec Kitty 3.x. Access doctrine artifacts programmatically via DoctrineService. Resolve agent profiles. Load action-scoped governance context iteratively, not all at once. Triggers: "interview for charter", "generate charter"…

Priivacy-ai/spec-kitty · 135 tokens

spec-kitty-runtime-next

Drive the canonical spec-kitty next --mission control loop for mission advancement. Load agent profiles at init, apply action-scoped doctrine context at each step boundary, and pull specific tactics/directives on demand. Triggers: "run the next step", "what should runtime do next", "advance the mission", "what is the…

Priivacy-ai/spec-kitty · 110 tokens

spec-kitty-mission-system

Understand how Spec Kitty missions work: the 4 built-in mission types, how they define workflows via step contracts and action indices, how missions and work packages relate, how templates are resolved through the 6-tier chain, and how doctrine artifacts (procedures, tactics, directives) compose mission behavior.…

Priivacy-ai/spec-kitty · 160 tokens