The Agent Skills Directory

Data Ingestion Surface: The skill processes output from external sources, including the ms CLI JSON responses and local ms.config.json files. While this is necessary for its functionality, it represents an indirect prompt injection surface where maliciously crafted application metadata or configuration values could attempt to influence the agent's behavior.
Ingestion Points: ms app list --json, ms app show --json, and the local ms.config.json file.
Capabilities: The skill utilizes the Bash tool to execute commands and the Read tool to access local configuration files.
Boundary Markers: The skill instructions do not explicitly define delimiters for external data, though the use of JSON formats provides inherent structural boundaries.
Command Execution: The skill executes shell commands via the ms CLI (e.g., ms auth status, ms app list). These commands are used to interact with the environment and retrieve application metadata, which is consistent with the skill's stated purpose of application discovery.

list-apps