The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content by opening user-provided URLs and executing JavaScript via run-code to inspect the page DOM. This creates a surface where a malicious webpage could attempt to influence the agent's behavior through embedded instructions.
Ingestion points: The URL provided via $ARGUMENTS is opened using playwright-cli open "$URL" in SKILL.md.
Boundary markers: There are no explicit delimiters or instructions to the agent to ignore potential commands found within the target website's content.
Capability inventory: The skill utilizes playwright-cli (network access, file writing for screenshots/snapshots/traces) and mkdir (filesystem access).
Sanitization: No sanitization or filtering of the webpage content is performed before it is processed by the agent's analysis routines.
[COMMAND_EXECUTION]: The skill uses several shell commands including playwright-cli, mkdir, and date to orchestrate tests and manage output directories. These commands are used for their intended purpose of automated testing and result organization.

test-responsive