The Agent Skills Directory

[COMMAND_EXECUTION]: The helper script scripts/with_server.py allows for the execution of arbitrary shell commands via the --server argument and a trailing command argument. It utilizes subprocess.Popen with shell=True to facilitate server startup (e.g., npm run dev), which is a sensitive operation if the command string is influenced by untrusted data.
[INDIRECT_PROMPT_INJECTION]: The skill's primary function is to ingest and analyze content from local web applications, creating a surface for indirect prompt injection. A malicious or compromised web application could include instructions within its DOM designed to manipulate the agent's behavior.
Ingestion points: Data from web application DOM content and console logs are read into the agent's context through Playwright's inspection capabilities, as demonstrated in examples/element_discovery.py and examples/console_logging.py.
Boundary markers: No explicit delimiters or instructions are provided to help the agent distinguish between application content and its own operational instructions.
Capability inventory: The agent can execute shell commands via scripts/with_server.py and generate/execute Python scripts using the Playwright library.
Sanitization: The skill does not implement sanitization or filtering for the data extracted from the browser environment before it is processed by the agent.

webapp-testing