The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to use playwright-cli (via npx) and curl to browse the web and fetch external documentation.
[EXTERNAL_DOWNLOADS]: At runtime, the skill may download the playwright-cli package using npx if it is not already present in the environment.
[PROMPT_INJECTION]: User-provided input is directly interpolated into the core execution instructions via the $ARGUMENTS placeholder.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) as it processes untrusted external data from the web.
Ingestion points: External documentation sites, GitHub repositories, and /llms.txt endpoints via playwright-cli and curl (SKILL.md).
Boundary markers: The skill lacks explicit delimiters or instructions to the agent to disregard commands embedded within the fetched external documentation.
Capability inventory: The skill has the ability to write to the local filesystem (research/web/) and execute shell commands (npx, curl).
Sanitization: There are no instructions provided to sanitize or validate the content retrieved from external URLs before processing.

explain-code