Back to skill
Skillv1.0.0
ClawScan security
Playwright CLI Automation · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignMar 12, 2026, 3:51 AM
- Verdict
- benign
- Confidence
- high
- Model
- gpt-5-mini
- Summary
- The skill is an instruction-only Playwright CLI guide whose requested actions and examples match its stated purpose; nothing in the bundle is disproportionate or clearly malicious, but exercise normal caution when installing packages and handling credentials.
- Guidance
- This skill is a plain instruction guide for using Playwright CLI and appears coherent with that purpose. Before using it: (1) verify Playwright is installed from official npm/pypi sources and run installs in a controlled environment; (2) inspect any generated scripts (codegen output) before executing; (3) avoid hard-coding real credentials — use secure secret storage or ephemeral test accounts; (4) treat saved storage files (auth.json) as sensitive; and (5) note the skill bundle has no homepage or source link, so if provenance matters, prefer official documentation from Microsoft Playwright or an owner you trust.
Review Dimensions
- Purpose & Capability
- noteThe name/description claim this is the Microsoft Playwright CLI and the SKILL.md provides installation commands and CLI usage that align with that purpose. However the skill bundle has no source/homepage and is instruction-only (no code or official links), so the package is a documentation/instruction wrapper rather than an officially published Playwright artifact — this is plausible but reduces provenance/trust.
- Instruction Scope
- noteThe instructions stay within the expected scope (install Playwright, install browsers, run screenshots, codegen, tests, and examples). They include examples that automate logins and suggest saving/loading storage files (auth.json) and embedding username/password placeholders — these are expected for browser automation but involve handling sensitive credentials and session files, so users should avoid storing secrets in plain files or run such scripts without review.
- Install Mechanism
- okThere is no install spec in the skill itself (instruction-only). The SKILL.md recommends standard install commands (npm install -g @playwright/test or pip install playwright and running playwright install) which are normal for Playwright. Note that those commands will download third-party code from public package registries; review and run them in a trusted environment.
- Credentials
- okThe skill declares no required environment variables, credentials, or config paths. The examples reference usernames/passwords and storage files but do not require any undeclared secrets. This is proportionate to a browser automation guide, but users should avoid embedding real secrets in scripts or saving auth JSON files without protection.
- Persistence & Privilege
- okThe skill does not request persistent presence (always is false), does not modify other skills or system-wide settings, and contains no install-time hooks. It only provides runtime instructions for the agent to follow when invoked.