The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from websites, which serves as an ingestion point for potential malicious instructions. * Ingestion Points: The skill captures the rendered DOM, raw source code, and website assets in steps 1, 2, and 4 (SKILL.md). * Boundary Markers: The instructions do not provide delimiters or 'ignore embedded instructions' warnings for the external content. * Capability Inventory: The agent is instructed to use wget for recursive crawling and performs deep parsing of headings, body copy, and links. * Sanitization: No explicit sanitization or filtering of external content is described.
[COMMAND_EXECUTION]: The skill provides a specific shell command for recursively downloading website assets. * Evidence: Section 2 of SKILL.md includes the command: wget -r -l 3 -k -p -E -nc https://example.com/page.

clone-website