playwright-e2e-init
Pass
Audited by Gen Agent Trust Hub on Apr 10, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes shell commands to install the @playwright/test package and browser binaries via bun and bunx.
- [EXTERNAL_DOWNLOADS]: During initialization, browser binaries are downloaded from official Playwright repositories.
- [PROMPT_INJECTION]: The skill analyzes project files to generate relevant test cases, which introduces an indirect prompt injection surface if the source code contains malicious instructions.
- Ingestion points: Application source files read by the agent (Next.js/React components).
- Boundary markers: Absent. No specific delimiters or instructions to ignore embedded content are defined.
- Capability inventory: File writing (test creation) and command execution (running tests).
- Sanitization: Absent. No validation or filtering of read content is specified.
Audit Metadata