playwright-recording
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill's content consists of legitimate technical documentation and code templates for browser automation. The recommended libraries and tools are industry-standard.
- [COMMAND_EXECUTION]: The reference material includes a utility template that uses child_process.execSync to invoke ffprobe for calculating video metadata. This is a common and appropriate use of shell execution for this context.
- [INDIRECT_PROMPT_INJECTION]: The templates facilitate visiting and interacting with external websites, which represents an indirect prompt injection surface. However, the skill does not perform automated decision-making based on untrusted web content. 1. Ingestion points: page.goto() calls in SKILL.md and reference.md. 2. Boundary markers: None present in code templates. 3. Capability inventory: File system operations (fs), shell command execution (ffmpeg/ffprobe), and browser automation control. 4. Sanitization: Not provided within the templates.
Audit Metadata