tour: Skill for Claude Code

.agents/skills/webapp-perf-cross-check/SKILL.md

webapp-perf-cross-check is a skill for Claude Code, Codex from a9a4k/tour. It costs 122 tokens per session (1,245 once invoked), scanned A, original, MIT.

A method for checking whether a web app is genuinely changing the browser page during React re-render cascades. React DevTools can count internal commits that cause no actual browser changes.

In plain words
What is it for?
Comparing React render data with a Chrome performance trace and a MutationObserver, which records actual page changes.
Why use it?
It prevents you from mistaking inflated React counters for real performance problems and making unnecessary fixes.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: installed under .agents/ (shared by several agents).

This is a9a4k/tour's own configuration. It tells Claude Code and Codex how to work on tour itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything tour configures →

View source ↗ a9a4k/tour
Reuse

Borrowing it

Nothing to install: this file belongs to a9a4k/tour. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/a9a4k/tour/main/.agents/skills/webapp-perf-cross-check/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/a9a4k/tour

Made for: Claude Code, Codex.

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 webapp-perf-cross-check

README.md
[![agentmods](https://agentmods.dev/badge/skills/a9a4k/tour/webapp-perf-cross-check/github.svg)](https://agentmods.dev/skills/a9a4k/tour/webapp-perf-cross-check)
Your own site
<a href="https://agentmods.dev/skills/a9a4k/tour/webapp-perf-cross-check"><img src="https://agentmods.dev/badge/skills/a9a4k/tour/webapp-perf-cross-check/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.

agentmods 80×15 button for webapp-perf-cross-check

Your own site · 80×15
<a href="https://agentmods.dev/skills/a9a4k/tour/webapp-perf-cross-check"><img src="https://agentmods.dev/badge/skills/a9a4k/tour/webapp-perf-cross-check.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 122 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,245 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00122 $0.01245
Opus 5 $0.00061 $0.00622
Sonnet 5 $0.00024 $0.00249
Haiku 4.5 $0.00012 $0.00125

Measured 9d ago against content hash c8c0b3db2a29, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

webapp-perf-cross-check 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 9d ago.

The scan reads SKILL.md. This mod also ships 1 executable file (scripts/aggregate-trace.py), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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/skills/webapp-perf-cross-check/SKILL.md · 114 lines

How it starts

The opening of the file, as written. The whole thing — 114 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Webapp perf cross-check

Prerequisites

The React-renders instrument requires agent-browser to be opened with React DevTools enabled:

agent-browser open <url> --enable react-devtools

Without this flag, react renders start fails with a non-obvious error about the DevTools hook not being installed. Chrome trace and MutationObserver work without it.

The trap

React DevTools' render profile reports a DOM column per component (e.g. 320/325). It counts fiber commits, not browser DOM mutations. A component can show "320 DOM updates" while the actual browser performed zero. React schedules a commit, the reconciler diffs the new tree against the old, finds nothing to update, but the profiler still ticks the counter.

Treat React DevTools render counts as a hypothesis, never as evidence of real cost. Confirm with browser-side signals before writing any patch.

Cross-check pattern

Run all three before deciding a cascade is real. agent-browser supports all three.

1. React renders (fiber commits — often inflated)

agent-browser react renders start
# ... do the action ...
agent-browser react renders stop --json

Note the Re-renders and DOM columns and the change reasons — they tell you which prop / state changed, which is the actual diagnostic value.

2. Chrome trace (real browser work)

agent-browser trace start
# ... do the action ...
agent-browser trace stop
# trace is saved under ~/.agent-browser/tmp/traces/trace-<ts>.json
python3 scripts/aggregate-trace.py <path-to-trace.json> "<label>"

scripts/aggregate-trace.py sums main-thread cost per event name and per EventDispatch:<type>, then prints the cross-check targets summary.

Targets per interaction:

  • EventDispatch:click / keydown < 50 ms → no long task
  • Layout count ≤ 2 per interaction (more = layout thrash)
  • Paint < 16 ms → frame budget intact

3. MutationObserver (real DOM mutations)

agent-browser eval "
  window.__mut = 0;
  const o = new MutationObserver((muts) => { window.__mut += muts.length });
  o.observe(document.body, {childList:true, subtree:true, attributes:true, characterData:true});
  'observing'
"
# ... do the action ...
agent-browser eval "window.__mut"

Read the full file on GitHub · 114 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 9d ago First seen · 114 lines · 122 tokens per session scan A c8c0b3db2a29

Subscribe to this mod's changes

webapp-perf-cross-check is a skill published in the GitHub repository a9a4k/tour (5 stars, last pushed 2mo ago), licensed MIT. It adds 122 tokens to every session and 1,245 once invoked, about $0.0006 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.