web-perf-audit

📖 English · 简体中文 · 繁體中文 · 日本語 · 한국어 · Español · Türkçe · Русский

Fixes that ship > advice that sits in a dashboard.
Multi-Agent Web Performance Audit Pipeline — SCAN → ANALYZE → SCORE → REPORT.
Not another Lighthouse wrapper. A complete audit-to-fix workflow with CWV impact estimates.
Works with Claude Code, Cursor, VSCode Copilot, and any CI pipeline.

Why web-perf-audit

You run Lighthouse. It says LCP 4.2s, score 62. You stare at the screen. Now what?

Every tool tells you what's wrong. None of them tell you what to change, line by line, and why that specific change will work. web-perf-audit fills that gap — it runs a multi-agent pipeline that:

Scans your page with Lighthouse + Puppeteer for raw CWV metrics
Analyzes across 5 dimensions in parallel (rendering path, resource hints, images, fonts, JS bundles)
Scores every finding against its real Core Web Vitals impact (not discovery order)
Generates a prioritized report with copy-pasteable fixes, each backlinked to the web.dev principle that explains it

"The goal isn't a 100/100 Lighthouse score — it's a page that loads fast and responds instantly for real users."

✨ Core Features

🎯 web.dev Standards-Aligned Every audit rule traces back to web.dev/learn/performance. LCP ≤ 2.5s, INP ≤ 200ms, CLS ≤ 0.1. Not our opinions — Google's thresholds, measured at p75.	🤖 Multi-Agent Audit Pipeline 5 specialized agents run in parallel: Project Scanner, Resource Analyzer, Metric Scorer, Change Predictor, and Optimization Planner. Each does one thing well.
🔧 Deterministic + Measured Hybrid Static checks (missing `defer`, wrong `font-display`, absent `srcset`) are deterministic — same page, same result. Runtime metrics (LCP, TBT) come from real browser measurement via Lighthouse + Puppeteer.	📊 Offline Dashboard Interactive Chart.js dashboard with score ring, CWV cards, category breakdown, and priority-ordered issue list. Works entirely offline — no external API calls after audit.
🔁 Incremental Git-Aware Audits SHA256 fingerprint caching skips unchanged pages. `perf-diff` compares two audits across branches — catch performance regressions before they merge, not after.	🧩 Cross-Platform & Multi-IDE Runs on macOS / Linux / Windows. Native Claude Code skill, plus Cursor rules, VSCode Copilot instructions. One-liner CI integration for GitHub Actions, GitLab CI, Jenkins.

🏗 Architecture

Multi-Agent Audit Pipeline

graph TB
    subgraph Trigger["🔔 Trigger"]
        CLI["/perf analyze URL"]
        CI["git push / PR opened"]
        IDE["Claude Code / Cursor / Copilot"]
    end

    subgraph Stage1["Stage 1 — SCAN"]
        Scanner["🔍 Project Scanner Agent<br/>· Framework detection (Vite/Webpack/Next)<br/>· Build artifact discovery<br/>· .webperfignore filtering"]
        LH["📡 Lighthouse + Puppeteer<br/>· LCP / CLS / INP / TBT / FCP<br/>· CrUX field data (PSI optional)"]
    end

    subgraph Stage2["Stage 2 — ANALYZE (5 agents parallel)"]
        CRP["📐 Critical Path Agent<br/>render-blocking CSS/JS<br/>request chain depth<br/>DOM size audit"]
        Hints["💡 Resource Hints Agent<br/>preload/prefetch/preconnect<br/>missing crossorigin<br/>sync script detection"]
        Img["🖼 Image Audit Agent<br/>WebP/AVIF detection<br/>srcset/sizes check<br/>LCP image optimization"]
        Font["🔤 Font Audit Agent<br/>font-display analysis<br/>woff2 format check<br/>subsetting audit"]
        JS["⚡ JS Bundle Agent<br/>Puppeteer Coverage API<br/>Long Task observer<br/>third-party attribution"]
    end

    subgraph Stage3["Stage 3 — SCORE"]
        Scorer["⚖️ Metric Scorer Agent<br/>· CWV impact weighting<br/>· severity × reach multiplier<br/>· P0 (≥20) / P1 (10-19) / P2 (<10)"]
    end

    subgraph Stage4["Stage 4 — REPORT"]
        Reporter["📝 Optimization Plan Agent<br/>· match fix templates<br/>· generate before/after code<br/>· estimate millisecond savings"]
        Output["📦 Output<br/>audit-report.json<br/>report.md<br/>dashboard.html"]
    end

    subgraph Diff["🔄 Incremental / Diff Mode"]
        Fingerprint["SHA256 fingerprint cache"]
        Predictor["🔮 Change Predictor Agent<br/>· Git diff → impact estimation<br/>· regression risk flagging"]
    end

    CLI --> Scanner & LH
    CI --> Scanner & LH
    IDE --> Scanner & LH
    Scanner & LH --> CRP & Hints & Img & Font & JS
    CRP & Hints & Img & Font & JS --> Scorer
    Scorer --> Reporter --> Output
    Output --> Fingerprint
    Fingerprint --> Predictor
    Predictor --> Output

Layered Design

Layer	Responsibility	Technology
Trigger	CLI / CI / IDE invocation	Claude Code Skill, GitHub Actions, Cursor rules
Scan	Raw data collection + project discovery	Lighthouse CLI, Puppeteer, PageSpeed Insights API
Analyze	5-dimension static + runtime analysis	Python 3.8+ (HTMLParser, urllib), Node.js (Puppeteer)
Score	CWV-weighted priority calculation	Python scoring engine (7-category × severity multipliers)
Generate	Report rendering + fix templating	Python (Markdown/JSON), Chart.js (dashboard HTML)
Diff	Incremental caching + cross-branch comparison	SHA256 fingerprinting, JSON structural diff

How we adapted Understand-Anything's pipeline

Understand-Anything pioneered the multi-agent code analysis pipeline — project-scanner → file-analyzer → architecture-analyzer → tour-builder → graph-reviewer. We adapted this pattern for web performance:

Understand-Anything	web-perf-audit	Adaptation
`project-scanner` (language/framework detection)	`Project Scanner Agent` (build tool detection, .webperfignore)	Code structure → page structure
`file-analyzer` (function/class extraction)	`Resource Analyzer Agent` (HTML/CSS/JS static audit)	AST nodes → DOM + CSSOM nodes
`architecture-analyzer` (layer tagging)	`Metric Scorer Agent` (CWV impact scoring)	Architecture layers → performance dimensions
`tour-builder` (guided walkthrough)	`Optimization Plan Agent` (fix generation)	Code tour → fix checklist
`graph-reviewer` (integrity check)	`Change Predictor Agent` (regression detection)	Schema validation → performance regression

🚀 Quick Start

4 steps from zero to audit report

# 1. Clone
git clone https://github.com/EVEDensity/web-perf-audit.git && cd web-perf-audit

# 2. Install dependencies
npm install -g lighthouse && npm install

# 3. Run your first audit
python scripts/fetch_metrics.py "https://example.com" --output-dir .web-perf
python scripts/analyze_critical_path.py .web-perf/metrics.json -o .web-perf/critical-path.json &
python scripts/check_resource_hints.py "https://example.com" -o .web-perf/resource-hints.json &
python scripts/audit_images.py "https://example.com" -o .web-perf/images.json &
python scripts/audit_fonts.py "https://example.com" -o .web-perf/fonts.json &
node scripts/audit_js_bundles.js "https://example.com" --output .web-perf/js-bundles.json &
wait
python scripts/score_and_report.py "https://example.com" --output-dir .web-perf --format all

# 4. Open the dashboard
open .web-perf/dashboard.html   # macOS
start .web-perf/dashboard.html  # Windows
xdg-open .web-perf/dashboard.html  # Linux

Or use the one-liner (Claude Code)

/plugin install web-perf-audit@EVEDensity/web-perf-audit
/perf analyze https://example.com

📦 Installation

macOS / Linux

# One-shot install
curl -fsSL https://raw.githubusercontent.com/EVEDensity/web-perf-audit/main/install.sh | bash

# Or step by step
git clone https://github.com/EVEDensity/web-perf-audit.git ~/.web-perf-audit
cd ~/.web-perf-audit
npm install -g lighthouse
npm install
echo 'alias perf-audit="python ~/.web-perf-audit/scripts/fetch_metrics.py"' >> ~/.bashrc

Windows (PowerShell)

# One-shot install
iwr -Uri https://raw.githubusercontent.com/EVEDensity/web-perf-audit/main/install.ps1 -OutFile install.ps1; ./install.ps1

# Or step by step
git clone https://github.com/EVEDensity/web-perf-audit.git $env:USERPROFILE\.web-perf-audit
cd $env:USERPROFILE\.web-perf-audit
npm install -g lighthouse
npm install

IDE Plugin Setup

IDE	How to enable
Claude Code	`/plugin install web-perf-audit@EVEDensity/web-perf-audit`
Cursor	Add `.cursorrules`: `@web-perf-audit analyze on save`
VSCode Copilot	Add to `.github/copilot-instructions.md`: `Use web-perf-audit for performance reviews`
Codex / Gemini CLI	Copy `SKILL.md` to your skills directory

Prerequisites

Dependency	Version	Required For
Python	≥ 3.8	All analysis + scoring scripts
Node.js	≥ 18	Lighthouse CLI + Puppeteer (JS audit)
Lighthouse	latest (`npm i -g lighthouse`)	CWV metric collection
Puppeteer	^22.0.0 (`npm i puppeteer`)	JS Coverage + Long Task API

📖 CLI Command Reference

All commands follow the /perf namespace. Each entry shows: purpose, example, flags, and output.

`/perf analyze` — Full Audit

Run the complete SCAN → ANALYZE → SCORE → REPORT pipeline against a URL.

python scripts/fetch_metrics.py <URL> --output-dir .web-perf
# ... (5 parallel analyzers) ...
python scripts/score_and_report.py <URL> --output-dir .web-perf --audience dev --format all

Flag	Default	Description
`--output-dir`	`.web-perf`	Where to write output files
`--audience`	`dev`	`dev` (full + code fixes) or `pm` (summary + scores only)
`--format`	`all`	`json`, `markdown`, `html`, or `all`
`--strategy`	`mobile`	Lighthouse emulation: `mobile` or `desktop`
`--psi-key`	—	PageSpeed Insights API key (for CrUX field data)
`--extra-lighthouse-flags`	—	Pass additional flags to Lighthouse CLI

Output: .web-perf/audit-report.json, report.md, dashboard.html

`/perf dashboard` — Visual Dashboard

Launch the interactive Chart.js dashboard without re-running the audit.

# Opens .web-perf/dashboard.html in your browser
open .web-perf/dashboard.html

# Or serve via HTTP for remote access
python -m http.server 8080 -d .web-perf
# Then open http://localhost:8080/dashboard.html

The dashboard includes:

Score ring (color-coded A→F)
CWV metric cards with good/needs-improvement/poor badges
Category breakdown bar chart (7 dimensions)
P0/P1/P2 issue list with fix snippets
vs-previous-audit comparison panel (automatic if fingerprint exists)

`/perf diff` — Git Change Impact

Compare two audit reports to catch performance regressions before they merge.

# Compare two audit snapshots
python scripts/diff_report.py \
  .web-perf/before/audit-report.json \
  .web-perf/after/audit-report.json \
  --format both \
  -o .web-perf/diff-report.json

Output: diff-report.json (structured) + diff-report.md (human-readable) with:

Per-metric delta (LCP: 3200→4100ms ❌)
Category score changes (JS: 72→58 ❌)
Resolved issues (✅ 3 fixed) vs new issues (⚠️ 2 introduced)
Regression severity: 🔴 critical / 🟡 warning / 🟢 improvement

CI integration example:

# .github/workflows/perf-diff.yml
name: Performance Diff
on: [pull_request]
jobs:
  perf-diff:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
        with:
          fetch-depth: 0
      - name: Audit base branch
        run: |
          git checkout ${{ github.base_ref }}
          python scripts/fetch_metrics.py "$STAGING_URL" --output-dir .web-perf/before
      - name: Audit PR branch
        run: |
          git checkout ${{ github.head_ref }}
          python scripts/fetch_metrics.py "$STAGING_URL" --output-dir .web-perf/after
      - name: Diff & Comment
        run: |
          python scripts/diff_report.py \
            .web-perf/before/audit-report.json \
            .web-perf/after/audit-report.json \
            --format markdown -o .web-perf/diff-report.md
          gh pr comment ${{ github.event.pull_request.number }} --body-file .web-perf/diff-report.md

`/perf explain` — Single File Deep Dive

Analyze one HTML/CSS/JS file in isolation. Useful for debugging a specific component or template.

python scripts/check_resource_hints.py <URL> -o .web-perf/resource-hints.json
# Then read the per-file breakdown in audit-report.json

`/perf resource` — Static Resource Batch Audit

Audit all images, fonts, and third-party scripts referenced by a page without running Lighthouse.

# Run only the static analysis scripts (no browser needed)
python scripts/audit_images.py "https://example.com" -o .web-perf/images.json
python scripts/audit_fonts.py "https://example.com" -o .web-perf/fonts.json

Use when:

You're iterating on image/asset changes and want fast feedback
Running in a headless CI environment without a display
Quick pre-flight before a full audit

`/perf chat` — Conversational Performance Consulting

Ask natural-language questions about your audit results (Claude Code native).

"Why is my LCP so high?" "Which of these P0 issues should I fix first?" "What's the estimated LCP improvement if I implement all image fixes?"

The Claude Code skill reads your latest audit-report.json and answers with context from the 8 web.dev reference modules.

💼 Real-World Use Cases

1. Local Frontend Project Audit

# Start your dev server
npm run dev &

# Audit the local build
python scripts/fetch_metrics.py "http://localhost:5173" --output-dir .web-perf
# ... (5 parallel analyzers) ...
python scripts/score_and_report.py "http://localhost:5173" --output-dir .web-perf --audience dev --format all

Typical findings on first run:

3-8 render-blocking resources to defer or inline
40-60% unused JavaScript (Coverage API)
Missing width/height on hero images (CLS impact)
font-display: swap not set on custom fonts

2. CI/CD Performance Gate

# .github/workflows/perf-gate.yml
name: Performance Gate
on: [push]
jobs:
  gate:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - name: Full Audit
        run: |
          npm install -g lighthouse && npm install
          URL="https://staging.example.com"
          OUT=".web-perf"
          python scripts/fetch_metrics.py "$URL" --output-dir "$OUT"
          for script in analyze_critical_path.py check_resource_hints.py audit_images.py audit_fonts.py; do
            python "scripts/$script" "$URL" -o "$OUT/$(basename $script .py).json" &
          done
          node scripts/audit_js_bundles.js "$URL" --output "$OUT/js-bundles.json" &
          wait
          python scripts/score_and_report.py "$URL" --output-dir "$OUT" --format all
      - name: Check Thresholds
        run: |
          # Fail if overall score < 70
          SCORE=$(python -c "import json; print(json.load(open('.web-perf/audit-report.json'))['overallScore']['overallScore'])")
          if [ "$SCORE" -lt 70 ]; then echo "Score $SCORE < 70 — failing gate"; exit 1; fi

3. Git post-commit Incremental Scan

# .git/hooks/post-commit (add this to your repo)
#!/bin/bash
PREV_FINGERPRINT=$(cat .web-perf/fingerprint.txt 2>/dev/null || echo "")

python scripts/fetch_metrics.py "http://localhost:5173" --output-dir .web-perf
# ... (parallel analyzers) ...
python scripts/score_and_report.py "http://localhost:5173" --output-dir .web-perf --format json

CURR_FINGERPRINT=$(cat .web-perf/fingerprint.txt)
if [ "$PREV_FINGERPRINT" != "$CURR_FINGERPRINT" ] && [ -n "$PREV_FINGERPRINT" ]; then
  echo "⚠️  Performance profile changed. Check .web-perf/dashboard.html"
fi

4. AgentHub Official Plugin Integration

web-perf-audit is designed as a first-class AgentHub plugin:

// AgentHub plugin registry entry
{
  "name": "web-perf-audit",
  "type": "official-plugin",
  "repo": "EVEDensity/web-perf-audit",
  "entry": "SKILL.md",
  "commands": ["perf-analyze", "perf-dashboard", "perf-diff"],
  "category": "performance",
  "requires": ["python>=3.8", "node>=18", "lighthouse"]
}

Once registered, AgentHub users can install it with a single command and run performance audits across all their projects.

🔧 Extension Development

Add a Custom Audit Rule

Create a new rule file in rules/:

# rules/custom_security_headers.py
def audit_security_headers(url, html_content):
    """
    Check for performance-relevant security headers.
    Returns: list of issue dicts with {type, description, severity, fix, ref}
    """
    issues = []
    # Your detection logic here
    return issues

Register it in score_and_report.py:

# In score_and_report.py, add to the analyzer registry
from rules.custom_security_headers import audit_security_headers
ANALYZERS['securityHeaders'] = audit_security_headers

Assign a weight (if it should affect the overall score):

WEIGHTS['securityHeaders'] = 0.03  # 3% of total score

Build a Custom Agent

Agents follow a standard interface — input JSON, output JSON + text:

# agents/my_agent.py
def analyze(input_data: dict) -> dict:
    """
    Args:
        input_data: {"url": str, "html": str, "metrics": dict}
    Returns:
        {"issues": [...], "score": float, "summary": str}
    """
    # Your agent logic
    return {"issues": [], "score": 100.0, "summary": ""}

if __name__ == "__main__":
    import sys, json
    data = json.loads(sys.stdin.read())
    result = analyze(data)
    print(json.dumps(result, indent=2))

Wire it into score_and_report.py's agent registry, and it runs in parallel with the built-in agents.

Customize the Dashboard

templates/dashboard.html uses Chart.js and CSS custom properties:

/* Override in your own build */
:root {
  --bg: #0f172a;           /* Dark background */
  --card-bg: #1e293b;      /* Card surface */
  --green: #22c55e;        /* Good score */
  --yellow: #eab308;       /* Needs improvement */
  --red: #ef4444;          /* Poor */
  --blue: #3b82f6;         /* Accent */
}

Change the 6 CSS variables for your brand. The Chart.js config is in render() — swap bar charts for radar, add trend lines, whatever fits.

Local LLM Integration (Ollama / vLLM)

For air-gapped or private deployments, swap the scoring LLM:

# In score_and_report.py, replace the LLM call
import requests

def score_with_local_llm(issue):
    resp = requests.post("http://localhost:11434/api/generate", json={
        "model": "llama3.1",
        "prompt": f"Rate this performance issue's severity (1-10): {issue['description']}",
        "stream": False
    })
    return resp.json()["response"]

# Replace the default `estimate_severity()` call
severity = score_with_local_llm(issue)

For vLLM, change the endpoint to http://localhost:8000/v1/completions.

❓ FAQ

How is this different from running Lighthouse?

Lighthouse gives you a score and an audit list. web-perf-audit:

Runs Lighthouse as stage 1 (data collection) only
Then runs 5 additional analysis passes Lighthouse doesn't do: resource hint quality scoring, font @font-face rule parsing, real JS Coverage via Puppeteer, third-party attribution, DOM depth audit
Scores and prioritizes all findings against CWV impact — not Lighthouse's "opportunity" order
Produces structured JSON for CI, not just a web UI
Supports incremental diff between two audits for PR gating
Every fix backlinks to a web.dev knowledge module explaining why

How long does a full audit take?

30–90 seconds for a typical page. Breakdown:

Lighthouse: 15–40s (depends on page complexity + network)
5 parallel analyzers: 10–30s (dominated by Puppeteer Coverage API)
Scoring + report generation: < 1s

For CI, consider caching the Lighthouse run or using PSI API (faster, no browser overhead).

Why is the audit report so large?

The JSON report includes per-file breakdowns from the JS Coverage audit — if your page loads 200+ JS chunks, the unused-bytes-per-file list adds up. Use .webperfignore to exclude third-party resources you can't control (analytics, ads, chat widgets). The Markdown report is filtered to P0/P1 issues only and is usually < 50KB.

I get "permission denied" on install.sh

chmod +x install.sh && ./install.sh

Or run it through bash directly: bash install.sh

The metrics seem off — LCP is reported lower than I see in DevTools

Lighthouse uses simulated throttling by default (4x CPU slowdown, 1.6 Mbps network). This is intentional — it reflects median mobile users. To match your DevTools experience, pass --strategy desktop --extra-lighthouse-flags "--throttling-method=provided".

Does this work with SPAs? (React / Vue / Svelte)

Yes. The Puppeteer-based JS audit waits for networkidle2 before collecting Coverage data, so dynamically loaded chunks are captured. For fully client-rendered apps:

Use --strategy desktop if that matches your user base
Set --extra-lighthouse-flags "--preset=desktop" for realistic LCP
The static analyzers work on the server-rendered HTML — for SPA shell pages, the DOM-based audits (images, hints) may be limited. Use the JS Coverage audit as your primary signal.

Can I run this on a private/intranet site?

Yes. Lighthouse supports http://localhost:* and internal IPs. For self-signed certificates:

python scripts/fetch_metrics.py "https://internal.company.com" \
  --extra-lighthouse-flags "--chrome-flags='--ignore-certificate-errors'"

How do I share the dashboard with my team?

Three options:

Static HTML: .web-perf/dashboard.html is self-contained — upload to any static host (S3, Netlify, GitHub Pages)
CI artifact: Upload .web-perf/ as a build artifact — teammates download and open locally
GitHub Pages: Push dashboard.html to gh-pages branch with a data embed

Can I plug in a private LLM?

Yes. See Extension Development → Local LLM Integration. The scoring engine's severity estimation is the only part that benefits from an LLM — everything else is deterministic. Replace the estimate_severity() function with your Ollama/vLLM/OpenAI-compatible endpoint.

What's the difference between PSI and local Lighthouse?

	Lighthouse CLI	PageSpeed Insights API
Data source	Simulated lab data	CrUX real-user field data + lab
Speed	15-40s	5-10s
API key	Not required	Required (free tier: 25k/day)
Accuracy	Lab — consistent, reproducible	Field — real user experience
Offline	✅	❌ (needs network)

Use PSI for production monitoring (real-user p75 data). Use Lighthouse CLI for local dev and CI (no rate limits, offline-capable).

🤝 Community & Contributing

How to Contribute

New audit rule — add a .md to references/ following the problem → detect → fix → benefit template, then wire it into the relevant analyzer script
New agent — create a script in scripts/ (Python or Node.js), produce JSON output, register it in score_and_report.py
Dashboard improvements — edit templates/dashboard.html, test with sample JSON
Bug report — open an issue with your audit-report.json and the URL tested

Issue Template

### URL tested
### Expected behavior
### Actual behavior
### audit-report.json snippet
### Lighthouse version (`lighthouse --version`)
### Node.js version (`node --version`)
### Python version (`python --version`)

PR Guidelines

Keep scripts self-contained — no new dependencies unless absolutely necessary
Python: standard library preferred, urllib for HTTP
Node: Puppeteer is the only allowed runtime dependency
All scripts must pass: python -m py_compile <script> or node --check <script>

Related Repositories

Project	Description
AgentHub	Multi-agent orchestration platform — web-perf-audit is an official plugin
MineWorld	WebGL voxel engine — the performance lessons that inspired this tool

⭐ Star History

📄 License

Built on web.dev/learn/performance — the Web Performance curriculum by Google Chrome.
Pipeline architecture inspired by Understand-Anything.
Part of the AgentHub ecosystem.

_{📖 English ·
简体中文 ·
繁體中文 ·
日本語 ·
한국어 ·
Español ·
Türkçe ·
Русский}

Name		Name	Last commit message	Last commit date
Latest commit History 18 Commits
.claude-plugin		.claude-plugin
docs		docs
references		references
scripts		scripts
templates		templates
.webperfignore		.webperfignore
LICENSE		LICENSE
README.md		README.md
README_es.md		README_es.md
README_ja.md		README_ja.md
README_ko.md		README_ko.md
README_ru.md		README_ru.md
README_tr.md		README_tr.md
README_zh-CN.md		README_zh-CN.md
README_zh-TW.md		README_zh-TW.md
SKILL.md		SKILL.md
install.ps1		install.ps1
install.sh		install.sh
package.json		package.json

Folders and files

Latest commit

History

Repository files navigation

web-perf-audit

Why web-perf-audit

✨ Core Features

🎯 web.dev Standards-Aligned

🤖 Multi-Agent Audit Pipeline

🔧 Deterministic + Measured Hybrid

📊 Offline Dashboard

🔁 Incremental Git-Aware Audits

🧩 Cross-Platform & Multi-IDE

🏗 Architecture

Multi-Agent Audit Pipeline

Layered Design

How we adapted Understand-Anything's pipeline

🚀 Quick Start

4 steps from zero to audit report

Or use the one-liner (Claude Code)

📦 Installation

macOS / Linux

Windows (PowerShell)

IDE Plugin Setup

Prerequisites

📖 CLI Command Reference

/perf analyze — Full Audit

/perf dashboard — Visual Dashboard

/perf diff — Git Change Impact

/perf explain — Single File Deep Dive

/perf resource — Static Resource Batch Audit

/perf chat — Conversational Performance Consulting

💼 Real-World Use Cases

1. Local Frontend Project Audit

2. CI/CD Performance Gate

3. Git post-commit Incremental Scan

4. AgentHub Official Plugin Integration

🔧 Extension Development

Add a Custom Audit Rule

Build a Custom Agent

Customize the Dashboard

Local LLM Integration (Ollama / vLLM)

❓ FAQ

🤝 Community & Contributing

How to Contribute

Issue Template

PR Guidelines

Related Repositories

⭐ Star History

📄 License

About

Topics

Resources

License

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Uh oh!

Contributors

Uh oh!

Languages

`/perf analyze` — Full Audit

`/perf dashboard` — Visual Dashboard

`/perf diff` — Git Change Impact

`/perf explain` — Single File Deep Dive

`/perf resource` — Static Resource Batch Audit

`/perf chat` — Conversational Performance Consulting

Packages