The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/with_server.py is designed to execute arbitrary shell commands. It uses subprocess.Popen with shell=True to run server commands provided via the --server argument and subprocess.run to execute the main command. This functionality allows for unrestricted command execution on the host system.
[PROMPT_INJECTION]: The skill facilitates the processing of untrusted data from web pages and browser console logs through Playwright in examples/element_discovery.py and examples/console_logging.py. This creates an indirect prompt injection surface where malicious instructions embedded in a web application could influence the agent's behavior, especially given its access to shell execution tools.
Ingestion points: page.content(), page.locator().all(), and console event listeners.
Boundary markers: None present to separate web data from agent instructions.
Capability inventory: Arbitrary shell execution via scripts/with_server.py and file system writes in examples/ scripts.
Sanitization: No sanitization or validation is performed on data retrieved from the browser context.

webapp-testing