Introduction

Monospace connects the systems your organization already runs and serves them through one governed API layer. Register a data source and Monospace exposes it four ways — a REST API, a typed TypeScript SDK, an MCP server, and the Studio. All four read and write live data under one consistent permissions model.

Teams build internal tools and customer-facing applications on it, inside one set of guardrails. Agents work the same way, acting on live systems rather than reading from a copy.

What is Monospace?

Monospace introspects each connected data source and maps it to collections, fields, and relations. The REST API, the SDK generator, the MCP server, and the Studio are all built on that model. Reads and writes pass through Monospace into the connected source systems.

A Monospace instance is a single organization. The organization holds workspaces, and each workspace has its own members, roles, and data sources. Data stays in the source system — Monospace federates at query time rather than replicating.

Why Monospace?

Unify Data Sources

Register a source and Monospace introspects it, maps it to collections and fields, and reads and writes it in place. Nothing is copied, and no ETL runs.

Relational databases have native connectors for PostgreSQL, MySQL, MariaDB, and SQL Server. Managed services include Supabase and Neon for PostgreSQL, and Amazon RDS for PostgreSQL, MySQL, and MariaDB. Other systems such as internal APIs and SaaS applications are reached through custom connectors. Downstream, every source behaves the same way.

Build with AI

Point the SDK generator at a workspace and it produces a typed TypeScript client for that schema. Agents work through the MCP server — a fixed set of tools for reading the schema, querying items, and writing changes. The official Agent Skills package teaches an agent the API surface and the SDK.

Each workspace has its own MCP endpoint, and an agent authenticates with an API key. It acts with exactly that key's permissions. The Studio Assistant uses the same tools, with the permissions of whoever is signed in.

Govern Access Centrally

Roles, policies, and entitlements are set once and resolved at query time for every caller, on every interface. Permissions apply at the collection, item, and field level. Administrative and security activity is recorded in the audit log.

Core Capabilities

Unified Data Access

Connect PostgreSQL, MySQL, MariaDB, and SQL Server, plus anything else reached through a custom connector. This includes Amazon RDS for PostgreSQL, MySQL, and MariaDB. Monospace queries each source in place — nothing is copied, and no ETL pipeline runs.

Relations work across sources — a collection in one database can relate to a collection in another. One request resolves both, joining systems that were never designed to talk to each other.

Human + AI Collaboration

The Studio interface enables business teams to work alongside developers on live enterprise data. AI systems access the same governed APIs, ensuring consistency and security across human and automated workflows.

Enterprise-Grade Control

• Role-based access control across all databases
• Workspace and organization-level governance
• Automatic schema introspection and mapping
• Audit trails and compliance features

Built for Scale

• Real-time query-level federation
• Workspace isolation for multi-tenant and client-agency setups
• Clustered deployment across multiple engine nodes
• Cost-efficient integration with your existing AI tools