Skip to main content
Glama
ARAVINDHRAJA123

sql-review-agent

by ARAVINDHRAJA123
OverviewSchemaRelated ServersScoreDiscussions
Python
Remote

SQL Review Agent

An agentic AI system that reviews BigQuery SQL queries before you run them — catching performance issues, estimating cost, and suggesting rewrites.

What it does

Paste a BigQuery SQL query and the agent:

  1. Static analysis — instantly flags SELECT *, missing partition filters, cartesian joins

  2. Schema fetch — reads partition keys, clustering fields, and row counts from BigQuery

  3. Cost estimate — dry-runs the query to get bytes scanned without executing it

  4. Rewrite — returns improved SQL with a plain-English explanation

  5. Severity rating — none / low / medium / high / critical

Related MCP server: bigquery-mcp

Architecture

The agent is modelled as a LangGraph state machine:

Agent graph

  • call_llm — sends messages + tools to Claude/Gemini

  • run_tools — dispatches get_table_schema, dry_run_sql, write_report

  • route — conditional edge: loop back if tool calls remain, stop when write_report is called

Three trigger modes

Mode

How

CLI

python agent.py --sql "SELECT * FROM ..."

Web UI

python server.py → open localhost:5001

MCP

Any MCP-compatible client (Claude Code, Cursor, Zed, OpenClaw)

Setup

git clone https://github.com/ARAVINDHRAJA123/sql-review-agent.git
cd sql-review-agent
python3 -m venv venv && source venv/bin/activate
pip install -r requirements.txt

export GCP_PROJECT=your-project
export BQ_LOCATION=asia-south1
export GEMINI_API_KEY=your-key     # free tier: aistudio.google.com
# or: export ANTHROPIC_API_KEY=your-key

gcloud auth application-default login

MCP (Claude Code)

claude mcp add -s user sql-review -- \
  /path/to/venv/bin/python /path/to/mcp_server.py

Tools available in any MCP client:

  • review_sql — full agentic review (LLM + BQ)

  • quick_check — instant static analysis, no LLM needed

CLI usage

python agent.py --sql "SELECT * FROM \`project.dataset.table\`"
python agent.py --file query.sql --verbose

Web UI

Web UI

python server.py
# open http://localhost:5001

Tests

pytest
pytest tests/test_tools.py -v

Project structure

sql-review-agent/
├── agent.py          ← raw tool-use loop (Claude + Gemini)
├── graph_agent.py    ← LangGraph state machine (drop-in replacement)
├── mcp_server.py     ← FastMCP server (review_sql + quick_check tools)
├── server.py         ← Flask web UI + JSON API
├── tools/
│   ├── bq_tools.py   ← dry_run, schema, metadata, read-only guard
│   └── sql_tools.py  ← static analysis, table extraction
├── tests/
│   ├── test_tools.py ← 24 unit tests
│   └── test_agent.py ← 7 unit tests
└── docs/
    ├── architecture_graph.svg  ← state machine diagram
    └── web_ui.png              ← Flask UI screenshot

Stack

Python · BigQuery · Claude API · Gemini API · LangGraph · FastMCP · Flask · pytest

This server cannot be installed

RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

Maintenance

ActivitySlowing
ResponsivenessNo issues

Resources

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Connectors

Related MCP Servers

  • bigquery-mcp

    vozzor
    A
    license
    Not graded
    quality
    C
    maintenance
    Enables AI assistants to query and analyze Google BigQuery data, including schema browsing, running queries, and comparing datasets through natural language.
    MIT

  • bigquery-mcp

    aicayzer
    A
    license
    Not graded
    quality
    D
    maintenance
    MCP server for secure BigQuery access across multiple Google Cloud projects, enabling querying, schema exploration, and data analysis with SQL validation and read-only controls.
    2
    MIT

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/ARAVINDHRAJA123/sql-review-agent'

If you have feedback or need assistance with the MCP directory API, please join our Discord server