The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the installation of Playwright packages via NPM (playwright and @playwright/mcp@latest). While these packages originate from a well-known organization, they represent external dependencies required for the skill's operation.\n- [COMMAND_EXECUTION]: The skill workflow involves executing several shell commands, including running a Node.js Playwright client script and invoking NPM for dependency management. These capabilities are consistent with the skill's purpose as a developer tool.\n- [PROMPT_INJECTION]: The skill is subject to indirect prompt injection (Category 8) due to its interaction with and ingestion of data from external web pages.\n
Ingestion points: Game state data is retrieved from the web page via the window.render_game_to_text function within web_game_playwright_client.js.\n
Boundary markers: There are no explicit boundary markers or instructions to ignore embedded commands within the ingested game state JSON.\n
Capability inventory: The agent has the capability to write files (fs.writeFileSync) and execute the Playwright script, which interacts with the browser and local filesystem.\n
Sanitization: The skill does not implement sanitization or validation of the game state payload before it is provided back to the agent context.

develop-web-game