test-exploratory

Pass

Audited by Gen Agent Trust Hub on Aug 25, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill uses npx --yes @playwright/cli@latest to fetch and run the Playwright CLI from the official Microsoft-maintained NPM registry. This is a standard practice for browser automation tasks and targets a well-known, trusted service.
  • [REMOTE_CODE_EXECUTION]: By invoking the Playwright CLI via npx, the skill executes remote code. However, the source is an official, versioned package from a trusted vendor (Microsoft), intended for the skill's primary purpose of web testing.
  • [COMMAND_EXECUTION]: The skill instructions include shell commands to initiate browser sessions and perform automated discovery. These commands are scoped to the playwright-cli and are used as intended for QA exploration.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection because it interacts with live, untrusted web applications and processes their content (accessibility trees, console logs, and network events). The skill mitigates this risk by emphasizing the use of isolated browser sessions (-s=explore-guest) and requiring human triage of findings rather than immediate automated obedience to site content.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 25, 2026, 05:56 AM
Security Audit — agent-trust-hub — test-exploratory