Playwright Scraper Skill

• Outcome: Review
• Latest audit: Jul 10, 2026 · 4:28 AM UTC
• Version: 1.2.0

Overview

This is a working web-scraping skill, but it should be reviewed because it is built around evading anti-bot protections and saving page captures with limited safeguards.

Install only for scraping you are authorized to perform. Do not use stealth mode to bypass access controls, Cloudflare challenges, CAPTCHAs, rate limits, or login barriers without explicit permission. Treat screenshots, saved HTML, and JSON output as sensitive data, store them carefully, and delete them when no longer needed. Pin and update Playwright, and keep browser sandboxing enabled where possible.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands

Findings (9)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 86% confidence
• Finding: The skill documentation explicitly instructs users to rely on environment variables such as SCREENSHOT_PATH, WAIT_TIME, HEADLESS, SAVE_HTML, and USER_AGENT, which indicates runtime capabilities that are not declared in permissions metadata. This creates a transparency and policy gap: operators may grant or run the skill without understanding that it reads configuration from the environment and can affect execution behavior or output destinations.

Missing User Warnings

Medium

• Confidence: 89% confidence
• Finding: The README explicitly promotes stealth scraping and anti-bot evasion techniques such as hiding navigator.webdriver and simulating human behavior, but it provides no warning about legal, terms-of-service, or abuse implications. In an agent skill context, this can normalize or encourage bypassing site protections, increasing the likelihood of unauthorized scraping or policy-violating automation.

Missing User Warnings

Medium

• Confidence: 85% confidence
• Finding: The README instructs users how to save screenshots and full HTML from scraped pages without warning that those artifacts may contain personal data, credentials, session content, or other sensitive information. In an automation skill, this omission increases the risk of over-collection, insecure retention, and accidental disclosure of scraped data.

Missing User Warnings

Medium

• Confidence: 89% confidence
• Finding: The skill advertises screenshot and HTML saving but does not warn users that running it may create local files containing page contents, session state artifacts, or other sensitive data. On authenticated or sensitive pages, those saved artifacts could persist credentials, personal data, or proprietary content on disk without informed user consent.

Missing User Warnings

Medium

• Confidence: 92% confidence
• Finding: The documentation promotes stealth techniques specifically intended to bypass anti-bot protections, including hiding automation markers and mimicking human behavior, without any warning about legal, compliance, privacy, or account-security implications. In context, this increases misuse risk because the skill is optimized for evading defensive controls on protected sites rather than ordinary browsing automation.

Missing User Warnings

Medium

• Confidence: 90% confidence
• Finding: The examples explicitly promote scraping an anti-bot-protected site and saving screenshots/HTML, but provide no warnings about authorization, privacy, retention of captured data, or terms-of-service and operational impact. In a scraping skill, this omission matters because the examples normalize potentially sensitive collection and anti-bot evasion, making misuse more likely by downstream users.

Missing User Warnings

Medium

• Confidence: 92% confidence
• Finding: The troubleshooting and performance tips suggest manual login workflows, delays to avoid blocking, and proxy rotation, which together can facilitate persistence against defensive controls without any safety framing. In the context of a Playwright scraping skill advertised for anti-bot-protected sites, these recommendations increase the likelihood of unauthorized scraping, account misuse, and collection of restricted content.

Unpinned Dependencies

Low

• Category: Supply Chain

Content

"author": "多米",
  "license": "MIT",
  "dependencies": {
    "playwright": "^1.40.0"
  }
}

• Confidence: 87% confidence
• Finding: "playwright": "^1.40.0"

Known Vulnerable Dependency: playwright==1.40.0 — 1 advisory(ies): CVE-2025-59288 (Playwright downloads and installs browsers without verifying the authenticity of)

High

• Category: Supply Chain
• Confidence: 98% confidence
• Finding: playwright==1.40.0

VirusTotal

64/64 vendors flagged this skill as clean.

Static analysis

No suspicious patterns detected.