The Agent Skills Directory

[COMMAND_EXECUTION]: The scripts/with_server.py script uses subprocess.Popen with shell=True to execute strings provided via the --server argument. This is intended to support complex shell initialization commands such as cd directory && npm run dev but provides a high-privilege execution surface.
[INDIRECT_PROMPT_INJECTION]: The skill's primary function involves using Playwright to ingest and interact with web application content, creating a significant indirect prompt injection surface.\n
Ingestion points: Instructions in SKILL.md and example scripts like element_discovery.py demonstrate capturing page content and locator metadata.\n
Boundary markers: The instructions lack explicit delimiters or warnings to treat web content as untrusted data.\n
Capability inventory: The skill has access to full shell command execution via the provided helper scripts and local file system writes for logs and screenshots.\n
Sanitization: No filtering or sanitization of the scraped web content is performed before the agent processes it.

webapp-testing