The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill prompts the user to execute npx ai-prisma-gen, which downloads the ai-prisma-gen package from the public npm registry at runtime.
[REMOTE_CODE_EXECUTION]: The use of npx to run a package from an unverified third-party source constitutes remote code execution, as the package code is fetched and executed directly in the user's environment.
[COMMAND_EXECUTION]: The skill's primary functionality relies on executing shell commands (npx ai-prisma-gen ...) which could be manipulated if input is not properly sanitized.
[CREDENTIALS_UNSAFE]: The skill explicitly requires the OPENAI_API_KEY environment variable to be set. While standard for AI-based tools, this sensitive credential is exposed to the environment where the unverified ai-prisma-gen package is executed.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by taking natural language descriptions and passing them to an LLM to generate code (Prisma schemas).
Ingestion points: User-provided English descriptions passed as arguments to the CLI tool.
Boundary markers: None identified in the skill instructions.
Capability inventory: Execution of shell commands via npx and network access (to OpenAI API).
Sanitization: No evidence of input sanitization or validation of the natural language description before it is processed.

prisma-gen