prisma-postgres
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions involve processing data from external sources and performing file system operations, representing an indirect injection surface.
- Ingestion points: Data retrieved from the Prisma Management API ("api.prisma.io/v1") and CLI outputs (referenced in "references/management-api.md" and "references/create-db-cli.md").
- Boundary markers: No specific delimiters or warnings for embedded instructions are provided for data ingested from the API.
- Capability inventory: The agent is instructed to execute shell commands via "npx" and update local ".env" files (referenced in "SKILL.md").
- Sanitization: The instructions do not include explicit sanitization or validation steps for the content received from external API endpoints.
- [COMMAND_EXECUTION]: The skill uses various command-line operations to manage databases and link local projects.
- Evidence: Use of "npx create-db@latest", "npx -y @prisma/cli@latest", and "prisma postgres link" for database provisioning and environment setup (referenced in "SKILL.md").
- [EXTERNAL_DOWNLOADS]: The skill relies on external software packages provided by the vendor.
- Evidence: Instructions include installing and running packages such as "create-db", "@prisma/management-api-sdk", and "@prisma/cli" from the NPM registry (referenced in "SKILL.md" and "references/management-api-sdk.md").
Audit Metadata