The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to perform terminal operations including installing packages via pipx and executing a Python script located at ~/.claude/skills/qa/scripts/smoke_test.py with various arguments.
[EXTERNAL_DOWNLOADS]: The skill requires downloading the playwright library and its associated browser binaries from public software registries.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection as it processes data from external websites (e.g., page titles and UI elements) to verify application health. Malicious content on a targeted page could be used to influence the agent's behavior during result interpretation. Evidence: Untrusted data enters the context via the Playwright browser session; no boundary markers or sanitization steps are documented; the agent possesses command execution and filesystem access capabilities.

qa