The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill facilitates the installation of official libraries from the Prisma organization.
Evidence: npm install @prisma/management-api-sdk and npm install create-db mentioned in references/management-api-sdk.md and references/create-db-cli.md respectively.
[REMOTE_CODE_EXECUTION]: The skill instructs the agent to run CLI tools via npx, which downloads and executes the latest versions of Prisma's official utility packages.
Evidence: npx create-db@latest and npx prisma studio in references/create-db-cli.md and references/console-and-connections.md.
[COMMAND_EXECUTION]: The skill uses shell commands to automate database setup and environment configuration.
Evidence: npx create-db@latest --env .env writes sensitive connection strings directly to a local environment file, which is a standard developer practice for secret management.
[INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface where untrusted data from API responses or CLI outputs could influence the agent's behavior.
Ingestion points: The agent processes output from npx create-db --json and responses from the Management API (https://api.prisma.io/v1).
Boundary markers: No specific delimiters or safety instructions are provided to the agent to treat external tool output as untrusted content.
Capability inventory: The skill has capabilities for network operations (REST API calls) and file writes (writing .env files via CLI).
Sanitization: Validation and sanitization are managed by the underlying Prisma CLI and SDK, rather than explicit instructions within the skill itself.

prisma-postgres