Pylar

- Create governed SQL views— define exactly what data agents can access across connected databases using Pylar's SQL IDE.
- Generate MCP tools from natural language— describe a data interaction in plain English and let Pylar's AI produce the corresponding tool.
- Publish tools to agent builders— deploy MCP tools to Claude Desktop, Cursor, LangGraph, Zapier, n8n, and similar platforms without redeploying agents.
- Monitor agent interactions with Evals— track errors, query patterns, and tool performance through the built-in observability dashboard.
- Join data across multiple sources— query and combine data from Snowflake, BigQuery, PostgreSQL, HubSpot, Salesforce, and other connected sources in a single view.

Build AI agents that interact with your data securely. Connect databases, create governed views, and deploy MCP tools to any agent builder.

Pylar is asecure data access layer for AI agentsthat enables interaction with structured data sources without requiring direct database access or complex API integrations.
- Data Sourcesconnect to Pylar (Snowflake, BigQuery, PostgreSQL, HubSpot, Salesforce, and more)
- SQL Viewis created to govern exactly what data agents can access
- MCP Toolsare built on the view—multiple tools for different use cases
- Tools publishto Agent Builders (Claude Desktop, Cursor, LangGraph, Zapier, Make, n8n, and more)
- Evalsmonitors all tool interactions for observability and optimization

- ✅Single Control Pane: Update views and tools without redeploying agents
- ✅No Raw Access: Agents only access data through your governed views
- ✅Unified Interface: One MCP endpoint for all data sources
- ✅Real-time Observability: Monitor all agent interactions with Evals

Create SQL views that define exactly what data agents can access. Views are the only access level—agents never get raw database access.

Describe what you want in natural language, and Pylar's AI generates MCP tools for your agents. No manual coding required.

Join data across multiple databases, warehouses, and business applications. Query Snowflake, BigQuery, PostgreSQL, HubSpot, Salesforce, and more—all in one place.

Monitor agent performance with the Evals dashboard. Track errors, query patterns, and optimize your tools based on real usage data.

Update views and tools without redeploying agents. Changes reflect immediately across all agent builders—Claude Desktop, Cursor, LangGraph, Zapier, and more.

Connect your databases and data sources to Pylar. Supported sources include:

- Databases: BigQuery, Snowflake, PostgreSQL, MySQL, Redshift, MotherDuck, Supabase, and more
- Business Apps: HubSpot, Salesforce, Google Sheets, and more

Use Pylar's SQL IDE to create governed views of your data. Join across multiple databases, filter sensitive data, and define exactly what agents can access.

Create MCP tools using AI or manually. Each tool defines how agents interact with your views.

Publish your tools and get your MCP credentials. Connect to any agent builder—no API hassles, no redeployment.

Comprehensive guides covering everything from connecting databases to monitoring with Evals:

- Making Connections- Connect your data sources
- Creating Data Views- Build governed SQL views
- Building MCP Tools- Create tools for your agents
- Publishing Tools- Deploy to agent builders
- Connecting Agent Builders- Integrate with Claude, Cursor, LangGraph, and more
- Evals- Monitor and optimize agent performance

20 real-world agent examples across different domains:

- Customer Support & Success (4 examples)
- Sales & Revenue (4 examples)
- Marketing (4 examples)
- Product (3 examples)
- Finance (3 examples)
- Operations (2 examples)

Get answers to common questions and troubleshooting help:

- FAQ- Frequently asked questions
- Troubleshooting- Common issues and solutions

- Support Email:support@pylar.ai
- Get Started:Sign up for Pylar
- Website:pylar.ai

MCP server for Apache AGE graph databases on PostgreSQL. 21 MCP tools — the most comprehensive Apache AGE MCP server (graph CRUD, Cypher queries, batch transactions, semantic search, Graph RAG, vis.js visualization, export/import) - F#/.NET — the only non-Python Apache AGE MCP server, installs as a single dotnet tool - Production-grade — BenchmarkDotNet-verified performance (cached queries in 62 ns, Cypher in 1 ms) - Open source — MIT license, published on NuGet - Documentation — full docs site at neftedollar.com/age-mcp

Local semantic search over documents (txt, md, pdf, docx, pptx, csv). Fully offline, multilingual, hybrid vector + keyword search via LanceDB. No API keys, no cloud.

A local MCP server that gives Claude Code, Cursor and Codex read access to your DynamoDB schema and items, with every write staged for human review.