SchemaGate 1.1.0 introduces column-level access control for LLMs, addressing a vulnerability where models could infer sensitive data from column names even if the table was restricted. The new feature allows users to define rules in a JSON file to hide specific columns from LLM prompts based on user roles. This prevents models from generating queries that access sensitive information like salaries or national IDs, improving data security for text-to-SQL applications. SchemaGate supports various databases and can be integrated as a Python library, CLI, or MCP server. AI
IMPACT Enhances data security for LLM-powered text-to-SQL applications by preventing sensitive column exposure.
RANK_REASON This is a product update for a data security tool that enhances LLM interactions with databases.
- Claude
- Cursor+
- Human Resources
- JSON
- LLM
- MCP
- Microsoft SQL Server
- MySQL
- Oracle
- PostgreSQL
- SchemaGate
- SQLite
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →