playwright-skill
Pass
Audited by Gen Agent Trust Hub on May 10, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses
execSyncinrun.jsto automate the installation of dependencies and browser binaries. - [EXTERNAL_DOWNLOADS]: Fetches the
playwrightlibrary and the Chromium browser from official registries (NPM and Microsoft) during initialization. - [COMMAND_EXECUTION]: Implements a dynamic execution pattern where agent-generated JavaScript is written to temporary files and executed via
require()to fulfill automation requests. - [PROMPT_INJECTION]: Identified an indirect prompt injection attack surface:
- Ingestion points: Web content is ingested via
page.goto()inSKILL.mdandlib/helpers.jsworkflows. - Boundary markers: No specific delimiters or safety instructions are applied to separate external page content from agent instructions.
- Capability inventory: The skill has broad capabilities including file system access, network operations, and full Node.js execution through
run.js. - Sanitization: External web data is processed without sanitization or filtering logic.
Audit Metadata