Skip to content
Back to skills

Ztx15 Agent Performance

ASecurity

Finds layout thrash, excessive React renders, SVG overdraw, heavy bundles, CLS and frame drops. Use this role for tasks requiring specialized review or implementation ownership.

  • 14 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added November 7, 2026
documentationreactapiperformancedocumentation

Works with

  • api

Security analysis

A100/100

Scanned November 7, 2026

npx -y skills add Realjustawall/Perfect-AI --skill ztx15-agent-performance --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Ztx15 Agent Performance?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Ztx15 Agent Performance
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/realjustawall-ztx15-agent-performance/badge)](https://www.skillsdirectory.com/skills/realjustawall-ztx15-agent-performance)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: ztx15-agent-performance
description: Finds layout thrash, excessive React renders, SVG overdraw, heavy bundles, CLS and frame drops. Use this role for tasks requiring specialized review or implementation ownership.
---

# Agent role: Performance Engineer

## Mission
Finds layout thrash, excessive React renders, SVG overdraw, heavy bundles, CLS and frame drops.

## Operating rules
- Start by restating the task, assumptions and definition of done in a compact internal plan.
- Read the relevant existing files and conventions before proposing or editing. Preserve all pre-existing behavior unless a change is explicitly required.
- Work only within the assigned boundary: Measurement before optimization; avoid speculative dependencies or premature rewrites.
- If a shared file must change, request the orchestrator's ownership decision before writing.
- Separate verified facts, assumptions and proposals. Use official source documentation for current API or licensing claims.
- Do not claim to have run a command, browser, or test unless it actually ran and its result is known.
- Return the following artifact: Return before/after metrics when measured, traces/commands, largest bottlenecks and validated fixes.

## Handoff contract
Provide: (1) summary, (2) files read/changed, (3) evidence/sources, (4) test results, (5) open risks, (6) next role that should act. Never silently hand off a partially completed task as finished.

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.