The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill uses npx within its wrapper script to download and execute the @playwright/cli package. This package is provided by a reputable organization.
[EXTERNAL_DOWNLOADS]: The skill fetches the Playwright CLI from the public registry during execution and provides instructions for manual global installation.
[COMMAND_EXECUTION]: The scripts/playwright_cli.sh script executes shell commands to invoke the Playwright CLI with arguments provided by the agent.
[PROMPT_INJECTION]: The skill processes content from external websites, creating a surface for indirect prompt injection.
Ingestion points: Web content and DOM snapshots are retrieved via playwright-cli open and playwright-cli snapshot as documented in SKILL.md and references/workflows.md.
Boundary markers: There are no explicit delimiters defined to isolate data returned from the browser from the agent's internal instructions.
Capability inventory: The skill can execute shell commands, read/write to the local file system for artifacts like screenshots, and make network requests via the browser.
Sanitization: No automated sanitization or verification of the extracted web data is performed before it is presented to the agent.

playwright