Governed SQL Views
Create secure, scoped data views using Pylar's SQL IDE. These views are the only level of access agents get—they never have raw database access.
Write standard SQL queries, join across multiple databases in a single query, and organize views within projects. Agents can only query through your defined views, ensuring complete security and governance.
SQL IDE
Built-in SQL editor where you select data sources and write queries with full autocompletion and syntax support.
Cross-Database Joins
Join data across multiple databases and warehouses in a single query—unify BigQuery, Snowflake, and PostgreSQL seamlessly.
Complete Isolation
Views are the only access level. AI agents never get direct database access, ensuring complete security and governance.
AI-Powered MCP Tool Creation
Create MCP (Model Context Protocol) tools from your SQL views using AI or manual configuration. Build multiple tools on a single view, each governing how agents interact with your data differently.
Use natural language prompts like 'create a tool to fetch error codes, events, and deal information' and Pylar's AI configures the tool. Or fine-tune tools manually for complete control. Test before publishing to ensure they work correctly.
AI-Assisted Creation
Use natural language prompts to create tools—Pylar's AI configures function names, descriptions, parameters, and SQL queries automatically.
Multiple Tools Per View
Create different tools on the same view for different use cases—one for filtering by date, another for customer segments.
Test Before Deploy
Test your tools before publishing to ensure they work correctly. Verify parameters, query execution, and results.
Deploy to Any Agent Builder
Publish once, connect to any agent builder—code-driven like LangGraph, no-code like Zapier, or AI-native like Claude Desktop. All controlled from one unified platform.
Get your MCP HTTP Stream URL and Bearer Token, then paste them into any agent builder. Update views or tools in Pylar, and changes reflect automatically everywhere—no redeployment needed.
One MCP Server
Single connection point for all agent builders. Publish your tools once and connect to any MCP-compatible platform with the same credentials.
Universal Compatibility
Works with code-driven builders (LangGraph, OpenAI), no-code tools (Zapier, Make), and AI-native platforms (Claude Desktop, Cursor).
Automatic Updates
Update views or tools in Pylar and changes reflect automatically in all connected builders—no code changes or redeployment required.
Evals & Observability
Get comprehensive insights into how AI agents interact with your views. Click 'Evals' to see detailed analytics that help you optimize your views and tools.
Track successful queries and identify failures, analyze error patterns, understand query shapes, and access full query logs for debugging. Use insights to iteratively improve your views and tool configurations.
Interaction Patterns
See exactly how agents are using your tools—track usage over time, identify popular tools, and understand agent behavior.
Error Analysis
Detailed breakdown of errors—what they are, when they occur, and why they failed. Identify patterns and fix issues quickly.
Query Shape & Logs
Understand the types of queries agents are making and access full query logs for debugging and analysis.
Datasources
Connect to multiple databases and warehouses simultaneously. Join data across different sources in a single query, giving you flexibility without complexity.
BigQuery
Enterprise data warehouse for fast analytics.
Redshift
AWS cloud data warehouse optimized for scale.
Snowflake
Cloud-native data platform for unified analytics.
MotherDuck
Serverless DuckDB-based analytics for small teams.
Supabase
Open-source Postgres backend for modern apps.
MySQL
The world's most popular open-source database.
PostgreSQL
Advanced open-source relational database system.
MS SQL Server
Microsoft's trusted relational data platform.
Agent Builders
Publish once, connect to any agent builder—code-driven like LangGraph, no-code like Zapier, or AI-native like Claude Desktop. All controlled from one unified platform.
OpenAI
Deploy MCP tools directly to GPT-4, GPT-4 Turbo, and custom assistants.
Claude
Connect your Pylar tools to Anthropic's Claude models via MCP.
Cursor
AI-powered code editor with MCP support for seamless Pylar integration.
Windsurf
AI coding assistant that supports MCP tools for data access.
VS Code
Connect Pylar tools to Visual Studio Code with MCP extension support.
LangGraph
Integrate Pylar views and tools into LangGraph workflows and agents.
n8n
Automate workflows with Pylar data connections in n8n pipelines.
Zapier
Connect Pylar to thousands of apps through Zapier integrations.
Make
Build powerful automations with Pylar data in Make workflows.
Agno
Deploy Pylar tools to Agno's agent platform for custom AI solutions.
Quick start
Learn how Pylar works: create data views, build MCP tools, and connect to agent builders.
Creating Data Views
SQL queries that define what data AI agents can access. Write SQL in Pylar's IDE, join across databases, and organize views in projects. Agents never get direct database access.
Core Concepts · Nov 1, 2025
Building MCP Tools
The interface between your data views and AI agents. Create tools from natural language prompts or configure manually. Test before publishing.
AI Tools · Nov 3, 2025
Connecting Agent Builders
Connect to any agent builder—LangGraph, Zapier, Claude Desktop, and more. Paste your MCP URL and token. Changes sync automatically.
Deploy · Nov 5, 2025
View all docs →