The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill relies on the @metabase/representations npm package for core functionality such as schema validation and specification extraction. Use of npx to run these tools involves downloading and executing code from the official Metabase scoped package on the public npm registry.
[COMMAND_EXECUTION]: The instructions direct the agent to execute several CLI commands using npx, including validate-schema, generate-entity-id, and generate-uuid. These commands are standard operations within the Metabase development workflow for maintaining data integrity and generating unique identifiers.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to read, understand, and edit Metabase representation YAML files, which serves as a potential surface for indirect prompt injection if the processed files contain adversarial content.
Ingestion points: The agent reads entity definition files (Cards, Actions, Transforms) from project directories such as collections/ and transforms/.
Boundary markers: The instructions do not specify the use of delimiters or data-processing guardrails to protect the agent from instructions embedded within the YAML content or code comments.
Capability inventory: The skill has access to file system tools (Read, Write, Edit) and shell execution (Bash), allowing it to modify project files and run validation scripts.
Sanitization: There is no mention of sanitizing external data; the skill relies on the vendor's schema validation tool to ensure format correctness rather than security filtering.

metabase-representation-format