# performance-auditor

performance-auditor — a subagent that hunts bottlenecks: algorithm cost, N+1 queries, bundle size, wasted re-renders. You get a prioritized report.

- subagent · `type=agent` · Claude Code, Codex, Cursor · ★ 1357 repo stars
- Author: qdhenry
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/qdhenry/Claude-Command-Suite/tree/HEAD/.claude/agents/performance-auditor.md
- Updated: 2026-03-01
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/agents/performance-auditor

## What it does

It works through five areas. Algorithms: how time and memory grow with input size, loops that repeat work, data structures that don't fit the access pattern, missing memoization. Database: N+1 patterns, absent indexes, joins that pull too much, connection pool settings. Frontend: what landed in the bundle, routes that could load on demand, components re-rendering for no reason, memory that never gets released. Backend: API response times, caching, blocked I/O. Network: payload weight, compression, CDN, batched calls.

Measurement comes first, through Bash: it hunts oversized build files, counts installed packages with `npm list`, finds large `.js` files on disk. Then it reads source with Read, Grep and Glob to spot repeated work and blocking calls. The result is a written audit: a score, a table of current numbers against targets, each issue with its location and a fix, and a queue split into now, next sprint, this quarter.

The tool set is Read, Grep, Glob and Bash. Nothing writes to disk, so every fix arrives as text and code samples you apply yourself. Sample measurement commands lean on npm, and a project without Node needs its own.

## Install

### Claude Code (native platform)

**Manually: put the files in place**
- File: `.claude/agents/performance-auditor.md` — For all projects use ~/.claude/agents/. Claude delegates by the description field or an @-mention

### Codex

**Manually: put the files in place**
- File: `.codex/agents/performance-auditor.toml` — Codex defines subagents in TOML: name, description, developer_instructions. Move the Markdown text into developer_instructions by hand. For all projects use ~/.codex/agents/

### Cursor

**Manually: put the files in place**
- File: `.cursor/agents/performance-auditor.md` — Cursor also reads .claude/agents/

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/qdhenry/Claude-Command-Suite/tree/HEAD/.claude/agents/performance-auditor.md

## When to use it
- Find N+1 queries before a release
- Check how a hot loop scales with input size
- See what inflated the bundle and what to split off
- Get a fix list ranked for the next sprint
- Trace memory that leaks in a component tree

## Example prompts
- Audit `api/orders.js` for query count and response time.
- Check the frontend: bundle size, lazy routes, extra re-renders.
- Write a performance report and rank the fixes by urgency.

## Similar
- [performance-analysis](https://affpapa.org/en/nevibecode/skills/rudolf-s-performance-analysis.md) — skill · ★ 587 repo stars
- [golang-performance](https://affpapa.org/en/nevibecode/skills/golang-performance.md) — skill · 44500 installs
- [python-performance-optimization](https://affpapa.org/en/nevibecode/skills/python-performance-optimization.md) — skill · 35675 installs
- [swiftui-performance](https://affpapa.org/en/nevibecode/skills/swiftui-performance.md) — skill · 5000 installs
- [CPU Performance Engineering](https://affpapa.org/en/nevibecode/mcp/cpu-performance-engineering.md) — MCP server · ★ 1160 repo stars
- [m10-performance](https://affpapa.org/en/nevibecode/skills/m10-performance.md) — skill · 4000 installs

## Other versions with the same name
- [performance-auditor · gupsammy](https://affpapa.org/en/nevibecode/agents/samarth-gupta-performance-auditor.md) — ★ 276 repo stars

From the same set: [qdhenry/claude-command-suite](https://affpapa.org/en/nevibecode/by/qdhenry/claude-command-suite.md) — 112 cards

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
