test-visual-regression

Pass

Audited by Gen Agent Trust Hub on Oct 7, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill downloads the official @playwright/cli package from the NPM registry via npx. This is an expected dependency for visual testing and originates from a well-known service provider.- [DYNAMIC_EXECUTION]: Uses npx to dynamically execute code from the @playwright/cli package, which involves running external code to facilitate browser automation and screenshot creation.- [COMMAND_EXECUTION]: Executes shell commands via npx and potentially the project's local test runner to perform UI regression tests and establish baselines.- [INDIRECT_PROMPT_INJECTION]: The skill processes external application UI states, viewports, and metadata to establish pixel-based baselines, which creates a potential surface for instructions embedded in the app's UI.
  • Ingestion points: Browser content and interface elements accessed during the 'Observe' phase (Phase 0 and 1).
  • Boundary markers: No specific delimiters or safety instructions are defined to separate ingested UI text from the agent's internal reasoning.
  • Capability inventory: File system access to write screenshots to .playwright-mcp/ and repository baseline directories, and execution of CLI tools via npx.
  • Sanitization: The instructions do not specify sanitization or filtering of text content found within the target application's user interface.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 7, 2026, 11:27 PM