azuresql-db-seed

Pass

Audited by Gen Agent Trust Hub on Sep 28, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • Command Execution: The skill utilizes docker exec to run sqlcmd and bcp utilities within a local container environment. This is a standard method for performing database administrative tasks and data loading in containerized development setups.
  • External Resource References: The instructions reference a private container registry and a Microsoft-owned registration link (aka.ms). These resources are provided by the vendor to facilitate access to the specific database engine image mentioned in the skill.
  • Package Management: The skill recommends the installation of well-known development libraries including mssql, pyodbc, and faker. These are established packages in the Node.js and Python ecosystems used for database connectivity and synthetic data generation.
  • Data Handling Considerations: The skill facilitates the ingestion and processing of external data files and synthetic data to populate database tables.
  • Ingestion points: Processes data from CSV files and programmatically generated content (SKILL.md, references/seed-snippets.md).
  • Boundary markers: While specific delimiters for input content are not defined, the skill emphasizes the use of idempotent scripts and structured data loading.
  • Capability inventory: Includes shell command execution via docker exec and database interactions through sqlcmd and bcp utilities (SKILL.md, references/seed-snippets.md).
  • Sanitization: The skill promotes secure coding practices by mandating the use of parameterized statements and explicitly advising against string concatenation to mitigate SQL injection risks.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 28, 2026, 08:19 PM