The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the leverage CLI to perform operations such as checking user authentication status via leverage whoami and creating new background agent sessions via leverage sessions new.
[INDIRECT_PROMPT_INJECTION]: The skill handles user-provided task descriptions and instructions which are passed to a remote cloud environment, creating an attack surface for indirect prompt injection.
Ingestion points: User input is collected for the session title, space name, and prompt content (task details and instructions) within the leverage sessions new command template.
Boundary markers: No specific boundary markers or 'ignore' instructions are used to encapsulate the user-provided strings.
Capability inventory: The skill executes shell commands via the leverage CLI and manages remote agent sessions.
Sanitization: The skill does not perform any sanitization or validation on the user-provided inputs before passing them to the CLI.
[EXTERNAL_DOWNLOADS]: The skill provides instructions to install the leverage-cloud-agents skill from the leverage-computer/skills repository using npx if the necessary CLI tools are missing.

background-agents