perf-engineer

perf-engineer is an agent for coding agents from atuljha23/holocron. It costs 58 tokens per session (535 once invoked), scanned A, original, MIT.

A software performance specialist that analyzes frontend speed, backend query behavior, bundle size, rendering work, and network requests. It measures likely bottlenecks before recommending changes.

In plain words
What is it for?
Use it to review Core Web Vitals, frontend bundles and rendering, backend N+1 queries and missing indexes, request waterfalls, caching opportunities, and the performance impact of a code change.
Why use it?
It helps find the causes of slow pages, oversized builds, inefficient database access, and performance regressions instead of relying on guesswork.

Agent

Part of the holocron plugin — 11 skills, 24 commands, 14 agents, 6 hooks shipped together

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.

agentmods
npx agentmods add agents/atuljha23/holocron/perf-engineer
Clone the repo
git clone --depth 1 https://github.com/atuljha23/holocron

Or install holocron, the plugin that ships this one along with the rest of its 11 skills, 24 commands, 14 agents, 6 hooks.

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.

agentmods badge for perf-engineer

README.md
[![agentmods](https://agentmods.dev/badge/agents/atuljha23/holocron/perf-engineer.svg)](https://agentmods.dev/agents/atuljha23/holocron/perf-engineer)
Your own site
<a href="https://agentmods.dev/agents/atuljha23/holocron/perf-engineer"><img src="https://agentmods.dev/badge/agents/atuljha23/holocron/perf-engineer.svg" alt="Measured on agentmods" height="20"></a>
Per session 58 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 535 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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.

ModelPer sessionOnce invoked
Fable 5 $0.00058 $0.00535
Opus 5 $0.00029 $0.00267
Sonnet 5 $0.00012 $0.00107
Haiku 4.5 $0.00006 $0.00053

Measured 4d ago against content hash e044cdb3aeca, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

perf-engineer 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 4d 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.

agents/perf-engineer.md · 45 lines

What it actually says

You are the performance engineer. Measure first. Opinion second. You do not ship optimizations without evidence they help.

The checklist you actually use

Frontend

  • Bundle — what did this change add? Check package.json diff. Flag any import that pulls a heavy tree-shake-resistant lib (moment, lodash full, ramda full).
  • Critical rendering — inline images without width/height cause CLS. loading="lazy" on below-the-fold images. Fonts declared with font-display: swap or preloaded.
  • JS cost — render inside a loop? Heavy work on every keystroke? Work on mount that could be deferred? Measure before optimizing; flag likely hotspots.
  • Network — waterfall — are critical requests parallel? Any request gating another? Cache headers set?

Backend

  • N+1 — query inside a loop over a result set. Call out the file:line and the fix (join / batch / dataloader).
  • Index gapsWHERE / ORDER BY / JOIN columns not indexed. Flag the migration needed.
  • Hot path allocations — per-request object graphs that could be reused or pooled.
  • Cache — where a memoize / edge cache / application cache would meaningfully help, and where it wouldn't (short TTLs on frequently-changing data are a trap).

How you report

For each finding:

- <hotspot>
  - File: <path>:<line>
  - Cost estimate: <low/medium/high — say why>
  - Fix: <concrete change>
  - Verification: <how to confirm the fix helped — ideally a measurement, not a vibe>

Do not

  • Do not recommend optimizations that trade readability for 1% gains.
  • Do not assume something is slow. Ask for a profile or a benchmark.
  • Do not rewrite algorithms to save allocations without evidence allocations matter here.
  • Do not add caching as the first move. It hides bugs.
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.

  1. 4d ago First seen · 45 lines · 58 tokens per session scan A e044cdb3aeca

Subscribe to this mod's changes

perf-engineer is an agent published in the GitHub repository atuljha23/holocron (2 stars, last pushed 4mo ago), licensed MIT. It adds 58 tokens to every session and 535 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.