AdvancedAll AIs

Analyze and optimize frontend performance

Analyze the performance issues of a frontend application and produce an optimization plan prioritized by gains/effort ratio, covering bundle, rendering, network and Core Web Vitals.

Paste in your AI

Paste this prompt in ChatGPT, Claude or Gemini and customize the variables in brackets.

Tu es un expert en performance web et optimisation frontend. Analyse les problèmes de performance de mon application et propose un plan d'action.

**Framework :** [FRAMEWORK] (ex: React, Vue, Next.js)
**Métriques actuelles Lighthouse :** [METRIQUES_LIGHTHOUSE] (LCP, FID/INP, CLS, TTFB)
**Contexte :** [CONTEXTE] (ex: e-commerce mobile-first, dashboard B2B)

**Description des problèmes observés :** [PROBLEMES_OBSERVES]

**Code ou configuration à analyser :**
```
[CODE_CONFIG]
```

Fournis une analyse en couches :
1. **Bundle JavaScript** : code splitting, lazy loading, tree shaking, taille des dépendances
2. **Rendu** : re-renders inutiles, hydration, SSR/SSG si applicable
3. **Réseau** : compression, cache, CDN, préchargement des ressources
4. **Images et assets** : formats modernes, lazy loading, responsive images
5. **Core Web Vitals** : actions spécifiques pour améliorer LCP, INP, CLS
6. **Plan d'action priorisé** : tableau gains/effort pour chaque optimisation
7. **Métriques cibles** : objectifs réalistes à atteindre

Why this prompt works

The layered structure (bundle/rendering/network/assets) reproduces the analysis method of a performance expert. The gains/effort table transforms a list of recommendations into an actionable plan. Realistic target metrics avoid impossible and frustrating objectives.

Use Cases

Improve the Lighthouse score of a landing pageReduce the LCP of an e-commerce siteOptimize a SPA for users on slow connection

Expected Output

5-layer analysis (bundle, rendering, network, assets, Core Web Vitals), action plan prioritized by gains/effort and realistic target metrics.

Learn more

Check the full skill on Prompt Guide to master this technique from A to Z.

View on Prompt Guide

Glossary Terms

Chain-of-thought Few-shot prompting Prompt injection Context Top-p (nucleus sampling)GPT Prompt leaking Hallucination