MCP For Claude Marketplace
SQL Parser AST Engine

SQL Parser AST Engine MCP Connector for Claude

A+

Parse any SQL query into a structured AST — extract tables, columns, JOINs, and WHERE clauses programmatically. Supports 15+ dialects including MySQL, PostgreSQL, and BigQuery. Your SQL injection firewall.

1 tools Official Updated Jun 28, 2026 Official Vinkius Partner
More Details Connect to Claude

A security agent receives a SQL query from user input. Is it safe? Does it access unauthorized tables? Is there a DROP TABLE hiding inside a subquery? An AI scanning the text will miss edge cases that a real parser catches.

This MCP parses SQL into a complete Abstract Syntax Tree — every table, column, JOIN, WHERE clause, subquery, and function call becomes a structured, inspectable object. Then it can rebuild valid SQL from the AST.

The Superpowers

  • SQL Injection Detection: Decompose any query to inspect for unauthorized operations, table access, and injection patterns.
  • 15+ Dialects: MySQL, PostgreSQL, MariaDB, SQLite, BigQuery, Snowflake, Hive, TransactSQL, and more.
  • Bidirectional: Parse SQL→AST and rebuild AST→SQL with full fidelity.
  • Table & Column Extraction: List every table and column referenced in a query — essential for data governance.
sql-parsingastquery-analysissecurity-firewalldatabase-securitycode-inspection

1 tools expose this connector's capabilities to your AI agent.

parse_sql

This is essential for security agents checking for SQL injection, DevOps agents auditing query performance, or any workflow that needs to understand SQL without executing it. Supported dialects: MySQL, PostgreSQL, MariaDB, SQLite, BigQuery, Snowflake, Hive, FlinkSQL, Noql, TransactSQL. Parses SQL queries into an AST and extracts tables, columns, and WHERE clauses. Supports 15+ dialects (MySQL, PostgreSQL, BigQuery, Snowflake, etc.)

See how to talk to your AI agent using SQL Parser AST Engine.

A user submitted this SQL query through our API. Parse it and check if it accesses any tables beyond 'orders' and 'products'.

Tables extracted: orders, products, users. ⚠️ 'users' table access detected — not in allowed list.

Extract all columns referenced in this BigQuery analytics query for our data governance audit.

14 columns extracted across 3 tables with full table.column qualified names.

Validate this PostgreSQL migration query for syntax errors before deploying to production.

AST parsed successfully — no syntax errors. Query is valid PostgreSQL.

Yes. Parse the query into AST and inspect the node types. If user input resulted in unexpected DROP, DELETE, UNION, or subquery nodes, the injection is exposed structurally — no regex guessing.

Related Connectors

MA

Mapbox MCP

27 tools Official

Integrate Mapbox mapping and location services into your AI agent—access vector/raster tiles, static images, directions, and manage datasets.

F View details →
KA

Kanye.Rest MCP

2 tools Official

Get random inspirational quotes from Kanye West directly in your AI agent via the Kanye.Rest API.

A+ View details →
AC

ActiveCampaign MCP

5 tools Official

Marketing automation and CRM — manage contacts, deals, lists, and automations via AI.

A+ View details →
ZA

Zapier MCP

5 tools Official

Monitor automated workflows, audit app connections, and search for Zap templates on Zapier — the leader in AI orchestration.

A+ View details →