← Back to Blog

iqtoolkit journal

🚀 IQToolkit Analyzer v0.2.8 & v0.3.0a6: Database-Direct Analysis & Multi-Platform Distribution

•5 min read•By Giovanni Martinez
releasepostgresqlmongodbdatabase performanceaidevtoolsquery optimizationdatabase analysisperformance tuninggoogle geminicli

Two major milestones for iqtoolkit-analyzer are here! v0.2.8 brings database-direct EXPLAIN analysis with contextual AI recommendations, while v0.3.0a6 delivers Google Gemini integration, MongoDB deep analysis, and multi-platform distribution across all major operating systems.

Watch demos and tutorials on the iqtoolkit YouTube channel:

youtube.com/@iqtoolkit

.

🔥 v0.2.8: Database-Direct EXPLAIN Analysis

The v0.2.8 release is a game-changer for PostgreSQL users. No more collecting log files—just point to your database and analyze queries instantly.

Major Features

🔥 Database-Direct EXPLAIN Analysis

Run EXPLAIN directly against live PostgreSQL databases:

iqtoolkit-analyzer \
  --config ~/.iqtoolkit/config.yaml \
  postgresql \
  --db-name my_database \
  --sql "SELECT * FROM orders WHERE customer_email = 'user@example.com'" \
  --output analysis.md

📝 IQToolkit Config Integration

Extended config file support with a databases:{} section for seamless database connections:

# ~/.iqtoolkit/config.yaml
default_provider: openai

providers:
  openai:
    api_key: ${OPENAI_API_KEY}
    model: gpt-4o-mini

databases:
  my_database:
    type: postgres
    host: localhost
    port: 5432
    database: production
    user: postgres
    password: ${DB_PASSWORD}

🎯 Contextual AI Recommendations

AI now provides performance-aware recommendations based on actual query execution time:

  • Fast queries (under 100ms): Acknowledges efficient execution
  • Moderate queries (100-1000ms): Balanced assessment
  • Slow queries (over 1000ms): Optimization suggestions

No more false-positive recommendations for already-optimized queries!

🔍 Full EXPLAIN Context

AI receives complete execution plan details:

  • Index Scan vs Sequential Scan analysis
  • Buffer usage statistics
  • Actual vs estimated rows comparison
  • Execution timing breakdown

What Changed

  • ✅ --db-name flag: Select database from config file
  • ✅ --sql flag: Analyze inline SQL queries
  • ✅ --query-file flag: Analyze SQL from files
  • ✅ Environment variable expansion in config files
  • 🐛 Fixed AI bias assuming all queries are slow
  • 🐛 Fixed EXPLAIN JSON parsing with leading whitespace

🚀 v0.3.0a6: Google Gemini + Multi-Platform Distribution

The v0.3.0a6 alpha brings massive improvements: native Google Gemini integration, MongoDB deep analysis, and installation via your favorite package manager.

🤖 Google Gemini AI Integration

Google Generative AI is now a first-class citizen in iqtoolkit-analyzer:

# PostgreSQL analysis with Google Gemini
iqtoolkit-analyzer postgresql \
  --db-name production \
  --use-ai \
  --ai-model gemini-2.0-flash

# MongoDB analysis with AI recommendations
iqtoolkit-analyzer mongodb \
  --connection mongodb://localhost:27017/mydb \
  --use-ai

New CLI flags for AI control:

  • --use-ai or --ai: Enable AI recommendations
  • --ai-model: Override default model (e.g., gemini-1.5-pro, gemini-2.0-flash)
  • --ai-stream: Enable streaming output for real-time feedback
  • --ai-key: Override API key (environment variable preferred)

Context-aware recommendations detect workload patterns (OLTP, OLAP, batch, reporting) and tailor suggestions accordingly.

🐘 PostgreSQL Deep Analysis

New analysis modules powered by pgtools:

  • pg_stat_statements analysis: Slow query detection and optimization
  • pg_stat_user_tables insights: Table-level performance metrics
  • pg_stat_activity real-time analysis: Long-running query detection, blocking sessions, idle-in-transaction detection
  • pg_locks contention detection: Lock analysis and deadlock prevention
  • Vacuum/autovacuum statistics: Dead tuple monitoring and maintenance recommendations

🍃 MongoDB Deep Analysis

New MongoDB analysis capabilities:

  • Profiler deep analysis: Execution efficiency, document scanning patterns
  • Index usage optimization: Missing index detection and usage recommendations
  • Aggregation pipeline analysis: Stage-by-stage optimization
  • currentOp analysis: Long-running operation detection, blocking operation identification, lock contention analysis

📦 Multi-Platform Distribution

Install iqtoolkit-analyzer on any platform with your preferred package manager:

Ubuntu/Debian:

sudo apt-get install iqtoolkit-analyzer

Fedora/RHEL/Rocky/AlmaLinux:

sudo dnf install iqtoolkit-analyzer

SUSE/openSUSE:

sudo zypper install iqtoolkit-analyzer

macOS (Homebrew):

brew install iqtoolkit/iqtoolkit/iqtoolkit-analyzer

Windows (Chocolatey):

choco install iqtoolkit-analyzer

Cross-platform (always works):

pip install iqtoolkit-analyzer
# or
pipx install iqtoolkit-analyzer

🐚 Shell Completion

Tab completion is now available for bash, zsh, and fish shells. Install from the completion/ directory and enjoy autocomplete for all commands and flags.

🔮 What's Coming in Full v0.3.0

The full v0.3.0 release is targeted for Feb-Mar 2026 and will include:

📋 Shared Contracts Package (iqtoolkit-contracts)

A standalone package with Pydantic models for clean architecture:

  • DatabaseConnection: Host, port, database, credentials
  • QueryInfo: Query text, duration, timestamp, source
  • ExplainPlan: Plan tree, execution time, rows, cost
  • AnalysisResult: Query, findings, recommendations, severity
  • AntiPattern: Name, description, impact, fix suggestion
  • AIRecommendation: Provider, model, recommendation, confidence

🏗️ Analyzer Service Refactor

  • Dependency injection for AI clients and database connections
  • Modular PostgreSQL and MongoDB analyzer modules
  • Enhanced testability with contract-based fixtures
  • EXPLAIN parser hardening for edge cases

🎨 CLI Interface Enhancement

  • Better contextual error explanations with suggested fixes
  • Comprehensive help text with examples
  • config validate command for configuration verification
  • --examples flag for each command

🤖 Future AI Providers (v0.4.0+)

  • Ollama integration: Local/self-hosted models
  • AWS Bedrock: Claude 3, Llama models with IAM management
  • Azure OpenAI: GPT-4, GPT-3.5 with deployment management
  • AI Provider Registry: Unified configuration with fallback chains

🎯 Who This Is For

DBAs and platform engineers who want instant database analysis without log file collection.

Developers who need AI-powered query optimization recommendations they can trust.

Teams running PostgreSQL or MongoDB who want a single tool for performance monitoring and tuning.

Enterprise users who need multi-cloud AI provider flexibility (Google Gemini, OpenAI, Ollama).

🚀 Get Started

Quick Install

# Latest stable (v0.2.8)
pip install iqtoolkit-analyzer

# Alpha with Google Gemini (v0.3.0a6)
pip install iqtoolkit-analyzer==0.3.0a6

First Analysis

# Set up your API key
export GOOGLE_API_KEY="your-gemini-api-key"

# Analyze a PostgreSQL query
iqtoolkit-analyzer postgresql \
  --host localhost \
  --database myapp \
  --sql "SELECT * FROM users WHERE created_at > NOW() - INTERVAL '7 days'" \
  --use-ai \
  --output analysis.md

Documentation

🧪 Feedback Welcome

I'm iterating quickly based on real-world usage. If you encounter issues, have feature requests, or want to share your experience:

  • Open an issue on GitHub
  • Join the discussions
  • DM me on social media

Your feedback shapes where this goes next!

Tags: #PostgreSQL #MongoDB #DatabasePerformance #AI #GoogleGemini #DevTools #IQToolkit #CLI