The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill instructs the agent to download and execute code from a public registry at runtime using npx -y luxin-cli@latest. This pattern executes the latest version of the vendor's package without local verification.
[EXTERNAL_DOWNLOADS]: The skill downloads and installs itself from a remote GitHub repository using npx skills add danielgwilson/luxin. It also fetches configuration and documentation from external URLs including luxin.sh/skill.md and luxin.sh/llms.txt.
[COMMAND_EXECUTION]: The installation and usage instructions involve shell commands that modify the environment, such as export HOME="$PWD/.agent-home" and mkdir -p, in addition to executing the CLI tool.
[PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection as it processes data returned from an external API and CLI tool.
Ingestion points: CLI command output and JSON responses from api.luxin.sh are incorporated into the agent's context.
Boundary markers: No explicit delimiters or instructions to ignore instructions embedded in the external data are present in the skill instructions.
Capability inventory: The skill possesses the capability to execute shell commands (npx), perform network operations, and modify the filesystem.
Sanitization: There is no evidence of sanitization or schema validation for the data received from the hosted API before it is processed by the agent.

image-to-3d