The Agent Skills Directory

[SAFE] (SAFE): The skill provides static documentation and code samples for the Prisma ORM. No malicious code, obfuscation, or persistence mechanisms were found.
[DATA_EXPOSURE] (SAFE): Code examples use process.env.DATABASE_URL for database connections, which aligns with security best practices for managing sensitive credentials.
[COMMAND_EXECUTION] (SAFE): The documentation for raw SQL queries ($queryRawUnsafe, $executeRawUnsafe) in references/raw-queries.md explicitly labels vulnerable patterns as unsafe and provides secure, parameterized alternatives. This educational approach reduces the risk of accidental SQL injection vulnerability introduction.
[PROMPT_INJECTION] (SAFE): No instructions were found that attempt to override agent constraints or safety filters.
[INDIRECT_PROMPT_INJECTION] (SAFE): While the skill reacts to user-provided data (queries about Prisma models), the evidence chain confirms its safety:
Ingestion points: User queries containing Prisma keywords (e.g., findMany, create).
Boundary markers: None present.
Capability inventory: The skill has no capabilities to execute shell commands, perform network requests, or modify the file system.
Sanitization: Not applicable as the skill only provides static documentation.

prisma-client-api