The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to use several shell commands to achieve its goals, including cp for file copying, find for directory searching, sed for text replacement, and ./gradlew for build verification. It also involves modifying the project's build.gradle.kts file to add dependencies.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it interpolates untrusted user input (such as package names and module names) directly into shell commands without defining sanitization or validation steps.
Ingestion points: User-supplied parameters for 'Target module', 'Package name', and 'Maximum Tuple size' are used to construct file paths and shell commands in SKILL.md.
Boundary markers: No delimiters or instructions to ignore embedded commands are used when processing these inputs.
Capability inventory: The skill utilizes powerful capabilities including full file system access (read/write) and arbitrary shell command execution via the agent's environment.
Sanitization: The instructions do not specify any validation or escaping logic for the user-provided strings before they are executed in the shell (e.g., within the sed commands in SKILL.md).

kotlin-tuple