skills/prisma/skills/prisma-orm-setup/Gen Agent Trust Hub

prisma-orm-setup

Pass

Audited by Gen Agent Trust Hub on Nov 9, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill analyzes project files such as package.json, lockfiles, and Prisma schemas, which are untrusted inputs that could contain malicious instructions.
  • Ingestion points: The agent is instructed to read the package manifest, lockfiles, and schema.prisma in SKILL.md (Step 1).
  • Boundary markers: There are no instructions for the agent to use delimiters or ignore instructions within these files.
  • Capability inventory: The skill can execute shell commands (prisma), install packages (npm install), and generate code (prisma generate).
  • Sanitization: No sanitization or validation of content from project files is specified.
  • [COMMAND_EXECUTION]: The skill uses several CLI commands to initialize and configure the environment.
  • Evidence: Commands such as prisma orm init, prisma skills sync, and npx prisma generate are utilized to manage the ORM lifecycle.
  • [EXTERNAL_DOWNLOADS]: The skill installs numerous Node.js packages to provide database connectivity and ORM functionality.
  • Evidence: It installs @prisma/client, prisma, and various drivers and adapters including @prisma/adapter-pg, @prisma/adapter-mariadb, @prisma/adapter-mssql, and @prisma/adapter-ppg. All packages are official Prisma components or well-known database drivers.
  • [DYNAMIC_EXECUTION]: The skill dynamically updates its own instruction set by syncing with the vendor CLI.
  • Evidence: The command prisma skills sync creates a prisma-8/SKILL.md file, which the agent is then directed to read and follow for further instructions (Step 4 in SKILL.md).
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 9, 2026, 08:44 PM