The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill uses npx -y xd-extract to download and run the xd-extract package. Because the package version is not pinned and the -y flag suppresses confirmation, there is a risk of executing malicious code if the package is compromised or updated with malicious intent.
[COMMAND_EXECUTION]: The skill executes shell commands using npx to run the inventory tool and open to view HTML contact sheets.
[EXTERNAL_DOWNLOADS]: The skill automatically downloads the xd-extract package from the npm registry at runtime.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external Adobe XD files and share links, which may contain attacker-controlled text or metadata designed to influence the agent's behavior.
Ingestion points: External Adobe XD share links (https://xd.adobe.com/view/...) and .xd files processed by the tool.
Boundary markers: No delimiters or instructions are used to separate the extracted design content from the agent's system prompt.
Capability inventory: The skill can execute shell commands (npx, open) and read from the file system.
Sanitization: There is no evidence of sanitization for the text elements (titles, layers, etc.) extracted from the design files before they are presented to the agent.

xd-inventory