The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill executes npx --yes getdesign@latest, which downloads and runs code from the npm registry at runtime. Because it uses the @latest tag, it will always execute the most recent version of the package without version pinning or integrity verification, making the system vulnerable to supply-chain attacks if the package is compromised.- [EXTERNAL_DOWNLOADS]: The skill initiates downloads from the npm registry and references external resources from github.com/VoltAgent/awesome-design-md. These sources are not part of the trusted vendor list and have not been audited for safety.- [COMMAND_EXECUTION]: The skill uses the npx command to interact with the project filesystem and execute external binaries. While intended for design automation, this provides a pathway for executing unverified code.- [INDIRECT_PROMPT_INJECTION]: The skill imports external DESIGN.md files that are explicitly intended to serve as the "source of truth" for the agent's future UI generation tasks, creating an attack surface where external content can influence agent behavior.
Ingestion points: SKILL.md describes a workflow to download external design templates into the local project via CLI.
Boundary markers: Absent. The agent is instructed to follow the imported files as a primary source of truth without safety delimiters.
Capability inventory: The skill can execute shell commands via npx and ls, and perform file-system write operations.
Sanitization: None. Imported templates are consumed by the agent without validation or filtering.

awesome-design-md