The Agent Skills Directory

[SAFE]: The skill acts as a command reference for the mcloud CLI tool, providing examples for listing, retrieving, and deleting projects. The operations are within the scope of the defined allowed-tools.
[COMMAND_EXECUTION]: The skill documents the use of mcloud subcommands and jq for project management. It follows safety best practices by explicitly flagging destructive operations (like projects delete) as irreversible and instructing the user to confirm IDs before execution.
[INDIRECT_PROMPT_INJECTION]: The skill defines patterns for processing project metadata (such as names and handles) returned from the Cloud API. While this involves ingesting external data, the processing is handled via standard CLI tools (jq), and the skill does not contain instructions that would cause the agent to execute untrusted content as code.

mcloud-projects