The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill provides instructions to download the Wardlock CLI installation script and binary via curl and go install from the developer's official GitHub repository and domain (CubieCode/wardlock and wardlock.nz).
[REMOTE_CODE_EXECUTION]: The installation process involves piping a remote script into the shell (curl ... | sh), which is a standard deployment method for the associated tool and originates from the developer's verified resource.
[COMMAND_EXECUTION]: The skill makes extensive use of the wardlock CLI to manage deployments, site policies, and sharing links. It specifies strict command formats, including the mandatory use of the --json flag for parseable output.
[INDIRECT_PROMPT_INJECTION]: The skill describes a workflow where content is downloaded from a remote server (wardlock pull) into a temporary directory for the agent to modify and re-publish. This creates a surface where instructions embedded in the remote files could potentially influence the agent's behavior during the editing phase.
Ingestion points: The wardlock pull command in the "Download a page" section downloads remote files into a local PULL_DIR for processing.
Boundary markers: The instructions do not specify the use of clear delimiters or instructions to ignore embedded content when the agent modifies the pulled files.
Capability inventory: The agent has the ability to read and edit local files, execute shell commands via the CLI, and upload modified content back to a remote server.
Sanitization: There is no explicit requirement for the agent to sanitize or validate the content pulled from the remote server before performing modifications.

wardlock-publish