The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/with_server.py facilitates the execution of arbitrary shell commands. It uses subprocess.Popen with shell=True to start server processes, which allows for shell features like command chaining (e.g., cd backend && python server.py).
Evidence: subprocess.Popen(server['cmd'], shell=True, ...) in scripts/with_server.py executes input provided via the --server argument.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) because it processes untrusted data from web applications.
Ingestion points: The agent is instructed to read HTML content (page.content()), inspect DOM elements (page.locator().all()), and capture console logs (page.on("console", ...)).
Boundary markers: No specific boundary markers or instruction-ignoring delimiters are suggested for the captured web content.
Capability inventory: The skill provides a environment capable of arbitrary command execution (scripts/with_server.py) and file system writes (examples/console_logging.py writing to /mnt/user-data/outputs/).
Sanitization: There is no evidence of sanitization or filtering applied to the web content before it is processed by the agent's logic.

webapp-testing