High-value integrations (Connectors) that extend Mattr's Agentic capabilities into your existing enterprise systems, focusing on deep data and workflow integration rather than simple API calls
For complex, multi-system automation, Mattr moves beyond basic API calls. This section documents the officially supported connectors that enable your Agents to seamlessly integrate with external services, databases, and enterprise control points (Management Control Points, or MCPs).
These connectors are crucial for both Workflow I (App-First) and Workflow II (Agent-First) scenarios where data persistence and system interaction are required.
Supported External Integrations
Mattr leverages structured protocols to enable intelligent interaction with external tools, allowing Agents to use real-world context and data from your connected systems.
Connector
Purpose
Mattr Workflow Use Case
Supabase
Serverless Backend & Database
Connects the App Builder output to a scalable, dedicated PostgreSQL database and backend. Used for permanent data persistence and external authentication.
GitHub
Source Control & DevOps
Provides secure Code Access & Git Integration. Allows Agents to read documentation (context) and perform actions like committing generated code or updating ticket statuses (Workflow II).
MCP Servers (Model Context Protocol)
Agentic Control Point
Enables Multi-Agent Orchestration and deep collaboration across complex enterprise systems. Used for connecting to private APIs, internal CRMs, and custom data sources.
Mechanism: API Call Skill vs. Connectors
The core API Call Skill provides generic HTTP communication, requiring manual setup of endpoint URLs, headers, and authentication.
In contrast, Enterprise Connectors (like the three listed above) utilize a dedicated, pre-configured framework. This framework manages authentication schemas (OAuth, Bearer Token) and endpoint protocols, ensuring secure, compliant, and reliable integration tailored for enterprise use.
Key Advantage: Using connectors ensures that the Agent can not only execute an API call (like the generic Skill) but can also read and write structured data within external tools (like updating a GitHub issue or pulling data from Supabase), ensuring the automation remains grounded in real-time context.
