webapp-testing

An agent skill by anthropics, from anthropics/skills. Tags: anthropic, debugging, developer-tools, web.

What it does

Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.

Install

With the skills CLI, which installs into Claude Code, Codex, Cursor and other agents:

npx skills add anthropics/skills --skill webapp-testing

Or copy the skill folder into Claude Code's skills directory by hand (~/.claude/skills for every project, or .claude/skills inside one):

git clone --depth 1 https://github.com/anthropics/skills
cp -r skills/skills/webapp-testing ~/.claude/skills/webapp-testing

Safety box score

Grade C, moderate risk: 23 of 100 risk points, rated 2026-10-20. The grade rates what the skill and its scripts can reach on the machine of whoever installs it, across eight categories from shell execution to secrets access. How the score works.

This skill is a local-browser automation toolkit built around Playwright plus one helper that starts servers for you. The main risk is that its bundled server wrapper can execute arbitrary shell commands, and the examples write screenshots/logs to fixed absolute paths outside the repo.

Category levels, from 0 (none) to 3 (high)
Category	Level	Why
Code & shell execution	3 (high)	`scripts/with_server.py` launches server commands with `subprocess.Popen(..., shell=True)`, so supplied server strings run as shell commands under the user account.
Network access	1 (low)	The examples drive a browser to `page.goto('http://localhost:5173')` and note `# Replace with your URL`, so the skill can generate HTTP traffic to whatever app is being tested.
Filesystem reach	1 (low)	The example scripts write to absolute paths outside the project, e.g. `/mnt/user-data/outputs/console.log`, `/mnt/user-data/outputs/static_page.png`, and `/tmp/page_discovery.png`.
Secrets & credentials	0 (none)	The files do not ask for passwords, tokens, browser profiles, keychains, or `.env` contents, and the scripts shown do not read credential stores.
Privilege & persistence	0 (none)	There is no `sudo`, global install, service registration, shell startup edit, or OS-level configuration in the shipped files.
Instruction hijack surface	1 (low)	`SKILL.md` tells the agent to use helper scripts as black boxes and says `DO NOT read the source until you try running the script first`, which pushes process rules that reduce review visibility.
Irreversible actions	0 (none)	The skill automates browser interaction and test steps, but it does not itself deploy, publish, send messages, book, pay, or delete anything.
Transparency	0 (none)	All shipped files are plain text Python/Markdown; there are no binaries, decoded-at-runtime blobs, or unpinned downloads in the reviewed skill.

Before installing

Read `scripts/with_server.py` first if you will use it; it executes server strings with `shell=True`.
Confirm that writing screenshots/logs to `/mnt/user-data/outputs` and `/tmp` is acceptable in your environment.
If you will point Playwright at anything beyond localhost, use a throwaway account/session first.

Rated 2026-10-20 by an automated review of 6 files from the skill's repository.

Source

Repository: anthropics/skills (all skills from this repository)
Path: skills/webapp-testing/SKILL.md
Branch: main
Updated: 2026-09-19

algorithmic-art — Creating algorithmic art using p5.js with seeded randomness and interactive parameter exploration. Use this when users request creating art using code.
claude-api — Reference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration.
frontend-design — Guidance for distinctive, intentional visual design when building new UI or reshaping an existing one.
mcp-builder — Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools.
skill-creator — Create new skills, modify and improve existing skills, and measure skill performance.
web-artifacts-builder — Suite of tools for creating elaborate, multi-component claude.ai HTML artifacts using modern frontend web technologies (React, Tailwind CSS.