← Back to Skills Marketplace
Playwright 1.0.3
by
howerlin0329
· GitHub ↗
· v1.0.0
· MIT-0
112
Downloads
0
Stars
2
Active Installs
1
Versions
Install in OpenClaw
/install playwright-1-0-3
Description
Browser automation via Playwright MCP. Navigate websites, click elements, fill forms, take screenshots, extract data, and debug real browser workflows. Use w...
Usage Guidance
This skill appears to be what it says: a Playwright/MCP browser automation guide. Before installing/using it, consider: (1) npx will download and execute npm packages at runtime — verify package names/versions and run in an isolated environment if you have concerns; (2) browser automation can visit arbitrary URLs and access content behind auth — do not provide high-privilege credentials unless necessary, and prefer scoped/temporary test accounts; (3) scraping can raise legal/terms-of-service or privacy issues — follow the guidance in scraping.md (respect robots, throttle, prefer APIs); (4) if you want to prevent autonomous agent runs, manage the agent's invocation permissions in your environment (autonomous invocation is platform-default but you can restrict when/what the agent may execute). Overall the skill is coherent, but exercise the usual caution when giving any automation tool network access or credentials.
Capability Assessment
Purpose & Capability
Name/description (Playwright browser automation, MCP, scraping, screenshots, tests) match the declared runtime needs (node, npx) and the content of SKILL.md. The guidance and examples reference Playwright and @playwright/mcp packages which are directly relevant; no unrelated credentials, binaries, or config paths are requested.
Instruction Scope
SKILL.md contains explicit instructions and code examples for launching browsers, navigating arbitrary URLs, interacting with pages, and extracting data — exactly what a Playwright automation skill should do. This is expected, but it also means the agent can access arbitrary web content (including authenticated pages if credentials are provided) and can run npx commands. The doc discourages abusive techniques (fingerprint hacks, rotating proxies) and recommends temporary auth state, which is good. Users should be aware that using the skill to scrape or automate sites can surface sensitive data if the target site requires authentication or if the agent is asked to operate on internal/private URLs.
Install Mechanism
This is instruction-only with no install spec in the registry, so nothing will be force-installed by the registry. The SKILL.md metadata suggests installing 'playwright' and '@playwright/mcp' via npm, which is reasonable and expected for this functionality. Note: runtime use of npx will fetch and run packages from npm on-demand — standard for Node tooling but worth awareness.
Credentials
The skill declares no required environment variables or credentials. Some example configs reference common CI envs (e.g., BASE_URL) and optional auth files for test suites — these are reasonable examples and not required by the skill. There are no unrelated secret requirements.
Persistence & Privilege
always:false and instruction-only design means the skill does not request permanent presence or elevated system privileges. The SKILL.md explicitly recommends keeping auth state temporary and not persisting browser sessions by default. It does not instruct modifying other skills or system-wide settings.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install playwright-1-0-3 - After installation, invoke the skill by name or use
/playwright-1-0-3 - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
- Clarified the MCP-first browser automation flow for more reliable action planning and execution.
- Improved quick-start guidance, especially for common form handling, screenshots, and data extraction tasks.
- Added practical guidance for Playwright MCP actions, direct scripts, tests, and selector strategies.
- Clearly differentiated use cases between static fetch, Playwright tests, MCP-driven browser control, and rendered-page scraping.
- Enhanced reference sections and core usage rules for easier onboarding and troubleshooting.
Metadata
Frequently Asked Questions
What is Playwright 1.0.3?
Browser automation via Playwright MCP. Navigate websites, click elements, fill forms, take screenshots, extract data, and debug real browser workflows. Use w... It is an AI Agent Skill for Claude Code / OpenClaw, with 112 downloads so far.
How do I install Playwright 1.0.3?
Run "/install playwright-1-0-3" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Playwright 1.0.3 free?
Yes, Playwright 1.0.3 is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Playwright 1.0.3 support?
Playwright 1.0.3 is cross-platform and runs anywhere OpenClaw / Claude Code is available (linux, darwin, win32).
Who created Playwright 1.0.3?
It is built and maintained by howerlin0329 (@howerlin0329); the current version is v1.0.0.
More Skills