Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Get Started Free →

bundle-size-analyzer · a5c-ai | DecimalAI

a5c-ai/bundle-size-analyzer

SkillScore82Maintained

Analyze and optimize application bundle size for desktop applications

Frontendby a5c-ai

Links this skill to your agents and keeps it current with the author’s updates.

Automatic

Frontend / UI pack

1 · One-time — install the CLI (docs ↗)

pip install decimalai

2 · Get the skill — writes .claude/skills/a5c-ai-bundle-size-analyzer/SKILL.md

decimalai skills pull a5c-ai-bundle-size-analyzer --out .claude/skills/

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/a5c-ai-bundle-size-analyzer/SKILL.md https://app.decimal.ai/s/a5c-ai-bundle-size-analyzer/SKILL.md

Raw file: /s/a5c-ai-bundle-size-analyzer/SKILL.md ↗ · machine-readable summary at /s/a5c-ai-bundle-size-analyzer.json ↗

In productionGlobal

how this skill behaves on real traffic · last 30 days

No production data for this skill yet — usage numbers appear here once agents start running it.

Anonymized aggregate across every org using this skill — proof from real production use, not a demo.

Does this skill help? · Benchmark: with vs. withoutsign in to run this on your agents

22 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 58d ago

Yes — +41% lift-15% tokens0% turns82% of whole cases pass with skill (18/22)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-01	✗→✓	▲ Improved	426%	0%
case-02	✗→✓	▲ Improved	2%	0%
case-05	✗→✓	▲ Improved	-44%	0%
case-08	✗→✓	▲ Improved	-52%	0%
case-09	✗→✓	▲ Improved	-36%	0%

Showing 5 of 22 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 7/31/2026

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

a5c-ai / babysitter

library/specializations/desktop-development/skills/bundle-size-analyzer/SKILL.md

View on GitHub ↗

License MITImported 7/31/2026Version v1

Skill contents

SKILL.md

bundle-size-analyzer

Analyze and optimize application bundle size to improve download times and memory usage.

Capabilities

Analyze bundle composition
Identify large dependencies
Detect duplicate packages
Suggest tree-shaking opportunities
Generate size reports
Track size over time

Input Schema

json
{
  "type": "object",
  "properties": {
    "projectPath": { "type": "string" },
    "bundler": { "enum": ["webpack", "vite", "rollup", "esbuild"] },
    "generateReport": { "type": "boolean", "default": true }
  },
  "required": ["projectPath"]
}

Webpack Bundle Analyzer

javascript
// webpack.config.js
const BundleAnalyzerPlugin = require('webpack-bundle-analyzer').BundleAnalyzerPlugin;

module.exports = {
  plugins: [
    new BundleAnalyzerPlugin({
      analyzerMode: 'static',
      reportFilename: 'bundle-report.html'
    })
  ]
};

Optimization Techniques

Tree shaking
Code splitting
Dynamic imports
Replace heavy dependencies
Compress assets

Size Targets

| App Type | Target | Acceptable | Too Large | |----------|--------|------------|-----------| | Simple utility | < 30MB | < 60MB | > 100MB | | Standard app | < 80MB | < 150MB | > 250MB | | Complex app | < 150MB | < 250MB | > 400MB |

Related Skills

startup-time-profiler
electron-builder-config

Illustrative Want to see what a5c-ai/bundle-size-analyzer would put in front of your model? Preview the rendered body and check its trigger phrases — no LLM call, no signup.

Latest benchmark · publisher runVerified

8/1/2026, 02:42 UTC · 22 cases

Lift

+41%

82% vs 41%

Agent turns

0%

1.0 vs 1.0

Tokens

-15%

1,685.05 vs 1,976.36

fail → pass · 9pass → pass · 9fail → fail · 4

Per-case results (22)

Case	Status	Duration (ms)			Turns			Tokens			Tool calls
Case	Status	Without	With	Δ	Without	With	Δ	Without	With	Δ	Without	With	Δ
case-01 Can you check my Vite project at `./client` to figure out which node modules are bloating the output build? I need an analysis of heavy dependencies and actionable tips to reduce the total download size.	fail→pass	4,509	10,321	+129%	1	1	0%	446	2,346	+426%	0	0	—
case-02 Please analyze the bundle for my standard web application under `./apps/web` configured with Webpack. Generate a size report document and evaluate whether our current bundle weight is within an acceptable target range.	fail→pass	16,446	12,907	-22%	1	1	0%	2,925	2,980	+2%	0	0	—
case-03 I have an Esbuild project in `./packages/cli`. Could you inspect its bundle composition, highlight any redundant or duplicate packages, and suggest dynamic imports or tree-shaking strategies to trim down its footprint?	fail→fail	14,325	12,193	-15%	1	1	0%	2,763	2,743	-1%	0	0	—
case-04 I want to add static HTML bundle visualizer reporting to my Webpack build configuration in `webpack.config.js`. Which plugin class and plugin options should be specified?	pass→pass	5,673	3,325	-41%	1	1	0%	1,150	1,042	-9%	0	0	—
case-05 We built a command-line utility project located at `./tools/cli` that generated a 45MB bundle output. Should this utility bundle size be considered acceptable, or does it exceed normal utility benchmarks?	fail→pass	13,287	6,186	-53%	1	1	0%	2,338	1,318	-44%	0	0	—
case-06 Our CLI utility tool under `./scripts/generator` produces a 120MB bundled artifact. Our team lead thinks anything under 200MB is fine for node CLI tools. Is 120MB acceptable for a simple utility?	pass→pass	14,193	7,055	-50%	1	1	0%	2,296	1,661	-28%	0	0	—
case-07 Our React front-end application at `./apps/dashboard` compiles into a single 280MB bundle. Is this size normal for a standard application, or has it breached acceptable bounds?	pass→pass	11,872	8,849	-25%	1	1	0%	2,025	2,025	0%	0	0	—
case-08 We are developing an enterprise suite under `./packages/enterprise` with extensive features. If our bundle is currently 110MB, does that meet the recommended size target for complex applications?	fail→pass	13,411	4,029	-70%	1	1	0%	2,356	1,136	-52%	0	0	—
case-09 Our complex enterprise application at `./apps/monolith` produces a 210MB bundle output. Is 210MB considered acceptable for a complex app?	fail→pass	13,427	6,327	-53%	1	1	0%	2,301	1,470	-36%	0	0	—
case-10 Our complex application at `./apps/desktop-suite` currently compiles to a 450MB output directory. Is this size acceptable for a complex app?	fail→pass	12,677	4,550	-64%	1	1	0%	2,210	1,197	-46%	0	0	—
case-11 Which bundlers are explicitly supported in the schema enum for bundle size analysis tasks?	pass→pass	6,241	1,445	-77%	1	1	0%	1,106	613	-45%	0	0	—
case-12 When configuring static reporting options for Webpack bundle visualization in `webpack.config.js`, what exact output report filename is designated?	fail→pass	4,344	1,514	-65%	1	1	0%	717	628	-12%	0	0	—
case-13 What five bundle size reduction strategies should be presented when reviewing an application footprint?	fail→pass	13,942	5,385	-61%	1	1	0%	2,402	1,315	-45%	0	0	—
case-14 When invoking a bundle size analysis request on `./services/frontend`, what is the schema default for the `generateReport` parameter?	pass→pass	8,500	1,590	-81%	1	1	0%	1,388	616	-56%	0	0	—
case-15 In the JSON schema for bundle size analyzer requests, which single property is required?	fail→pass	10,078	1,566	-84%	1	1	0%	1,722	624	-64%	0	0	—
case-16 Our Rollup build at `./packages/ui-kit` has multiple versions of `lodash` bundled in the build output. How should we identify duplicate packages and reduce bundle bloat?	pass→pass	13,492	9,196	-32%	1	1	0%	2,665	2,247	-16%	0	0	—
case-17 In our project at `./apps/marketing`, we import full library icon sets like `import * as Icons from 'lucide-react'`. What optimization technique should be applied here?	pass→pass	9,690	4,958	-49%	1	1	0%	1,831	1,230	-33%	0	0	—
case-18 Our admin portal at `./apps/admin` loads heavy charting libraries on the initial page load, causing high bundle download times. What architectural optimization technique reduces initial load size?	pass→pass	8,500	6,770	-20%	1	1	0%	1,577	1,661	+5%	0	0	—
case-19 After tree shaking and code splitting our project at `./web-app`, our raw JS and SVG assets are still taking up significant download bandwidth. What final optimization technique should be applied?	pass→pass	9,954	8,647	-13%	1	1	0%	1,638	1,820	+11%	0	0	—
case-20 We want to profile the runtime CPU execution time and V8 engine JIT parse/compile latency during cold start of our Node process at `./server.js`. Can you generate a CPU flamegraph profile?	fail→fail	14,644	11,178	-24%	1	1	0%	2,589	2,302	-11%	0	0	—
case-21 Can you configure `electron-builder.yaml` for `./desktop-app` to set up macOS code signing certificates, notarization API keys, and Windows NSIS installer targets?	fail→fail	12,912	14,912	+15%	1	1	0%	2,638	3,676	+39%	0	0	—
case-22 We need to set up Cloudflare CDN cache-control headers (`s-maxage=31536000, immutable`) and HTTP edge caching rules for our static web assets in NGINX. Can you draft the cache header configuration?	fail→fail	12,594	10,197	-19%	1	1	0%	2,397	2,421	+1%	0	0	—

These cases were also screened by the skill’s intent / safety pass.See the Trust review ↗Reproduce this benchmark in Colab ↗

How this skill was measured

DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +41 percentage points is the difference between those two pass rates over the 22 comparable cases.

Model measured: gemini-3.6-flash
Graded by: gemini-3.6-flash
Grading method: judged
Verification: verified
Pass rate without the skill: 40.9%
Pass rate with the skill: 81.8%
Cases attempted: 22
Cases counted toward lift: 22
Cases that regressed: 0
Skill version measured: v1
Run completed: 8/1/2026, 02:42 UTC

Measured example — same prompt, with vs. without the skill · from the benchmark run · case case-01

Prompt

Can you check my Vite project at `./client` to figure out which node modules are bloating the output build? I need an analysis of heavy dependencies and actionable tips to reduce the total download size.

Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.

Show both answers

SkillScore blends 4 signals1 of 4 complete

BenchmarkPass rate WITH skill on a verified run82

Live evalNeeds ≥10 evaluated traces in 30 days

AI ratingNeeds ≥10 rated traces

AdoptionNeeds ≥10 serves across ≥2 orgs in 30 days

Efficiency · separate axis (at equal correctness)+0% turns-18% tokens

Shown for transparency; not credited unless a verified run shows no correctness regression.

Backed by 1 of 4 signals — each added signal makes the score more robust.

Verified on gemini-3.6-flash · re-tested 58 days ago

Other measured skills in the registry, with their headline benchmark lift.