The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the user or agent to install the midjourney-pp-cli and a corresponding MCP server using package managers. Specifically, it uses npx to fetch @mvanhorn/printing-press-library and go install to fetch the module from github.com/mvanhorn/printing-press-library. These resources are owned by the vendor.
[COMMAND_EXECUTION]: The skill facilitates the execution of the midjourney-pp-cli binary to perform various Midjourney-related tasks. It passes user-provided arguments and subcommands directly to the shell environment using the Read Bash tool.
[CREDENTIALS_UNSAFE]: To function, the skill requires the user to set the MIDJOURNEY_COOKIE_HEADER environment variable. This variable contains sensitive authentication data used to communicate with the Midjourney API. The skill also manages configuration profiles and feedback logs stored in the ~/.midjourney-pp-cli/ directory.
[DATA_EXFILTRATION]: The CLI tool contains built-in features for sending data to external endpoints. The --deliver webhook:<url> flag allows users to route the output of any command to a remote URL. Additionally, the feedback command can be configured to automatically transmit local logs to a remote server defined by MIDJOURNEY_FEEDBACK_ENDPOINT.
[PROMPT_INJECTION]: The skill exposes an attack surface for indirect prompt injection by taking natural language input from the user and passing it to the midjourney-pp-cli which command to resolve functionality.
Ingestion points: User queries for capabilities in the Direct Use section.
Boundary markers: None present for the which command arguments.
Capability inventory: The skill can execute shell commands, perform network requests, and write to the local filesystem.
Sanitization: No explicit sanitization or validation of the natural language query is described.

pp-midjourney