The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted content from Minecraft datapack files (JSON configurations and .mcfunction scripts) during the editing and debugging process. While the skill includes a validator, it lacks specific prompt boundary markers to prevent the agent from potentially being influenced by instructions embedded within the datapack data.
Ingestion points: The agent reads and iterates over files within the data/ directory and pack.mcmeta metadata file.
Boundary markers: No explicit delimitation or instructions to ignore embedded commands are present in the skill's instructions.
Capability inventory: The skill allows for local file system inspection, directory traversal, and execution of validation scripts.
Sanitization: The skill provides basic syntax and structural validation through the validate-datapack.sh script.
[COMMAND_EXECUTION]: The skill includes local utility scripts (scripts/validate-datapack.sh and scripts/jq-shim.mjs) used to automate structural and syntax checks. These scripts perform routine file system operations such as directory searching, file existence checks, and JSON parsing using standard shell commands and Node.js built-ins.

minecraft-datapack