A set of rules for improving website loading speed, responsiveness, accessibility, and Lighthouse scores. Lighthouse is a tool that measures these areas in a web page.

Use it to investigate Core Web Vitals, image and font loading, JavaScript bundle size, API performance, layout shifts, and inefficient data queries.

It keeps optimization work tied to measured problems and avoids adding code changes that do not produce a real improvement.

performance — Cursor rule for Cursor by virastack/ai

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.

Clone the repo

git clone --depth 1 https://github.com/virastack/ai

Made for: Cursor.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/rules/virastack/ai/performance/github.svg)](https://agentmods.dev/rules/virastack/ai/performance)

Your own site

<a href="https://agentmods.dev/rules/virastack/ai/performance"><img src="https://agentmods.dev/badge/rules/virastack/ai/performance/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/rules/virastack/ai/performance"><img src="https://agentmods.dev/badge/rules/virastack/ai/performance.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 0 Nothing until a file matches its globs; then the whole rule loads.

When invoked 790 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

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.

Model	Per session	Once invoked
Fable 5.1	$0.00000	$0.00790
Opus 5.5	$0.00000	$0.00316
Sonnet 5.5	$0.00000	$0.00158
Haiku 4.5	$0.00000	$0.00079

Measured 1mo ago against content hash eba853b400f6, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

performance 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 1mo ago.

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.

templates/core/.cursor/rules/performance.mdc · 68 lines

How it starts

The opening of the file, as written. The whole thing — 68 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Performance & Accessibility (Lighthouse 95+)

Constraints

Measurement First: Avoid premature optimization. Identify specific bottlenecks via profiling (DevTools Performance tab) before applying any fixes.
Keep or Revert: If an optimization (useMemo, useCallback, lazy loading) does not show a measurable improvement, REVERT IT. Neutral changes add maintenance burden.
Log Attempts: Document failed performance attempts to prevent re-applying the same ineffective optimizations.
Performance Budgets:
- Core Web Vitals: LCP ≤ 2.5s, INP ≤ 200ms, CLS ≤ 0.1.
- API & Load: API response time < 200ms (p95). Time to Interactive (TTI) < 3.5s on 4G connections.
- Bundle Size: Keep initial feature chunks < 50kb. Lazy load heavy dependencies.
LCP Optimization: Prioritize loading the first visible hero assets or images.
Image & Data Standards:
- Define explicit width and height to prevent CLS.
- Use <picture> with source tags for complex art direction or resolution switching.
- Prevent N+1 query patterns; strictly use pagination or optimized includes/joins for all data-driven components.
Icon & Font Strategy: Use tree-shakable icons (e.g., lucide-react). Apply swap display strategy for fonts.
A11y SSOT: All interactive elements MUST have aria-label, role, or an associated <label>. Maintain WCAG AA compliance (4.5:1 contrast ratio). Color alone must never convey state — pair with text, icons, or patterns.
UI Delivery: Interaction, focus, responsive breakpoints, and state completeness are enforced in ui-components.mdc (Delivery Checklist). This file owns budgets, assets, and CLS/LCP.
Asset Links: Reference ui-components.mdc for Skeleton, focus, and layout implementation rules.

Bans

Layout Shift: Forbidden to render content without reserved dimensions (use Skeletons).
Native Img: Forbidden to use raw unoptimized <img> tags for hero images.
Blocking JS: Forbidden to use heavy 3rd party scripts in the main thread; defer them.
A11y Gaps: Forbidden to use div as a button without tabIndex and onKeyDown.
Alt Text: Forbidden to leave alt attributes empty or generic (e.g., "image").

Read the full file on GitHub · 68 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.

1mo ago First seen · 68 lines · 0 tokens per session scan A eba853b400f6

Subscribe to this mod's changes

performance is a cursor rule published in the GitHub repository virastack/ai (23 stars, last pushed 2mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 790 tokens. 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.

CSS and styling Claude Code Frontend agent-skills antigravity cursor skills virastack

marker-triage

Triage a Marker.io website-feedback backlog at scale. Use when asked to go through, categorize, prioritize, deduplicate, or make sense of many Marker.io issues at once. Splits content fixes from functional bugs, groups by page, rates criticality, flags duplicates and repeat reporters, surfaces quick wins, lists…

marker-io/mcp-skills · 85 tokens

marker-client-report

Turn a project's completed Marker.io work into a client-ready report. Use when asked to write a weekly or monthly recap for a client, summarize what was shipped or fixed this period, or produce a status update for a website project from Marker.io. Read-only: it reads issues and writes a markdown report, it does not…

marker-io/mcp-skills · 67 tokens

marker-insights

Aggregate Marker.io website feedback into recurring patterns. Use when asked to find the most common issues on a site, which pages get the most feedback, recurring content problems, or the content-vs-bug split across many Marker.io reports, and produce a prioritized brief for a website team or agency. Read-only: it…

marker-io/mcp-skills · 73 tokens

performance