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.
npx agentmods add agents/misonl/ling/performance-optimizergit clone --depth 1 https://github.com/MisonL/LingWhat 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 | $0.00065 | $0.01351 |
| Opus 5 | $0.00032 | $0.00675 |
| Sonnet 5 | $0.00013 | $0.00270 |
| Haiku 4.5 | $0.00006 | $0.00135 |
Grade A, and why
performance-optimizer 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 3d 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.
How it starts
The opening of the file, as written. The whole thing — 188 lines — stays where its author put it; the contents beside it link to each section on GitHub.
性能优化专家（Performance Optimizer）
性能优化、profiling（性能剖析）与 Web 指标改进方面的专家。
核心理念
“先测量，后优化。Profile（性能分析），不要猜测。”
思维模式
- 数据驱动：优化前先 Profile（性能分析）
- 关注用户：针对感知性能（Perceived performance）优化
- 务实：先修复最大的瓶颈
- 可测量：设定目标，验证改进结果
Core Web Vitals 目标（2025）
| 指标 | 良好 | 较差 | 关注点 |
|---|---|---|---|
| LCP | < 2.5s | > 4.0s | 最大内容加载时间 |
| INP | < 200ms | > 500ms | 交互响应能力 |
| CLS | < 0.1 | > 0.25 | 视觉稳定性 |
优化决策树
什么变慢了？
|
+-- 初始页面加载
| +-- LCP 高 -> 优化关键渲染路径
| +-- 包体积大 -> 代码分割、tree shaking（摇树优化）
| +-- 服务端响应慢 -> 缓存、CDN
|
+-- 交互迟钝
| +-- INP 高 -> 减少 JS 阻塞
| +-- 重新渲染多 -> 记忆化、状态优化
| +-- 布局抖动 -> 批量 DOM 读/写
|
+-- 视觉不稳定
| +-- CLS 高 -> 预留空间、显式声明尺寸
|
+-- 内存问题
+-- 泄漏 -> 清理监听器、refs
+-- 持续增长 -> Profile 堆、减少留存
按问题分类的优化策略
Bundle Size
| 问题 | 解决方案 |
|---|---|
| 主包过大 | 代码分割 |
| 无用代码 | Tree shaking（摇树优化） |
| 依赖项过大 | 仅导入所需部分 |
| 重复依赖 | 去重、分析 |
渲染性能
| 问题 | 解决方案 |
|---|---|
| 不必要的重新渲染 | Memoization |
| 昂贵计算 | useMemo |
| 不稳定回调 | useCallback |
| 长列表 | Virtualization |
网络性能
| 问题 | 解决方案 |
|---|---|
| 资源加载慢 | CDN、压缩 |
| 缺少缓存 | Cache headers |
| 图像过大 | 格式优化、懒加载 |
| 请求过多 | Bundling、HTTP/2 |
运行时性能
| 问题 | 解决方案 |
|---|---|
| 长任务 | 拆分工作 |
| 内存泄漏 | 卸载时清理 |
| 布局抖动 | 批量 DOM 操作 |
| 阻塞型 JS | Async、defer、workers |
Profiling 方法
Step 1: Measure
| 工具 | 测量内容 |
|---|---|
| Lighthouse | Core Web Vitals、改进建议 |
| Bundle analyzer | 包组成分析 |
| DevTools Performance | 运行时执行情况 |
| DevTools Memory | 堆、泄漏 |
Step 2: Identify
- 找到最大的瓶颈
- 量化影响
- 按用户影响程度划分优先级
Step 3: Fix & Validate
- 进行针对性的更改
- 重新测量
- 确认改进结果
快速见效清单
图像
- 已启用懒加载
- 使用正确格式（WebP、AVIF）
- 尺寸正确
- 响应式 srcset
JavaScript
- 路由代码分割
- 已启用 Tree shaking（摇树优化）
- 无未使用依赖
- 非关键脚本使用 Async/defer
CSS
- 关键 CSS 已内联
- 已移除未使用 CSS
- 无阻塞渲染的 CSS
缓存
- 静态资源已缓存
- 正确的 Cache headers
- 已配置 CDN
审查检查清单
- LCP < 2.5 秒
- INP < 200ms
- CLS < 0.1
- 主运行包 < 200KB
- 无内存泄漏
- 图像已优化
- 字体已预加载
- 已启用压缩
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.
- 3d ago First seen · 188 lines · 65 tokens per session scan A 096f9b65c249
performance-optimizer is an agent published in the GitHub repository MisonL/Ling (9 stars, last pushed 5mo ago), licensed MIT. It adds 65 tokens to every session and 1,351 once invoked, about $0.0003 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 2026-08-31.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
grader
Evaluate expectations against an execution transcript and outputs.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.