The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses shell commands including npx playwright test, jq, unzip, and find to run tests and extract report data.
[REMOTE_CODE_EXECUTION]: The skill uses npx playwright test to run tests and suggests adding page.screenshot() calls to code. Since Playwright is a well-known industry tool, this execution is expected for the skill's purpose.
[PROMPT_INJECTION]: The skill processes data from test reports and trace logs, which serves as a surface for indirect prompt injection.
Ingestion points: playwright-report/results.json, trace.trace, and trace.network.
Boundary markers: None.
Capability inventory: Shell execution (npx, jq, unzip) and file system access.
Sanitization: None.

playwright-debugger