The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill includes a static analysis tool, scripts/schema_validator.py, which processes external data in the form of schema files.
Ingestion points: The script reads files matching patterns like **/prisma/schema.prisma or **/drizzle/*.ts within a user-specified directory path provided via command-line arguments.
Boundary markers: None identified; the script reads the raw text of schema files.
Capability inventory: The script performs local file system reads and directory traversal (globbing). It does not perform network operations, file writes, or command execution.
Sanitization: The script uses regex patterns to extract model and enum definitions for naming and structure validation. It does not execute or evaluate the content of the files it reads.

database-design