The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the wget utility to download website resources, including recursive crawling of the target domain.
Evidence: Step 2 in SKILL.md instructs the execution of wget -r -l 3 -k -p -E -nc on user-provided URLs.
[INDIRECT_PROMPT_INJECTION]: The skill ingests and parses external website content, which could contain malicious instructions designed to manipulate the agent's behavior during the cloning process.
Ingestion points: External website DOM, raw source code, and metadata are captured via browser automation and wget as described in Steps 1 and 2 of SKILL.md.
Boundary markers: Absent. The skill does not provide specific instructions to the agent to treat external content as data only or to ignore instructions embedded within the source page.
Capability inventory: The skill utilizes shell command execution (wget), browser automation tools, and interaction with design tool APIs.
Sanitization: Absent. The instructions in Step 4 of SKILL.md encourage direct parsing of headings, body copy, and asset URLs from the captured DOM without validation or sanitization steps.

clone-website