Agent Framework
Overview
The Mastra Agent Framework is a Node.js-based AI orchestration platform designed to transform raw educational data into actionable insights through multi-agent collaboration. Built on the Mastra framework, it provides production-ready workflows for analyzing student wellbeing data, moderating educational content, and generating compassionate, evidence-based recommendations.
What It Does
The platform serves two primary domains:
-
K12 Safety Documentation — AI-powered editing and generation of Emergency Operation Plans (EOP) functional annexes and Quick Reference Guides (QRG) for K12 educational institutions, helping schools create and maintain critical safety documentation.
-
Content Moderation (TAP) — Analyzes text and image content for safety concerns using Azure AI Content Safety, performing automated batch tagging and content safety analysis for educational social media platforms.
-
Reasoning Engine — A shared, domain-agnostic analysis pipeline that accepts data from any source format (JSON, CSV, API payloads) and produces insights in any target format through pluggable intake/output adapters.
Key Capabilities
-
Multi-Agent Orchestration — Specialized AI agents collaborate through defined workflows, each contributing domain expertise (temporal analysis, pattern recognition, correlation detection, explanation generation, validation).
-
Production Database Integration — Full SQL Server persistence for workflow execution logs, agent outputs, prompt overrides, and large payload storage with content-addressable deduplication.
-
Format-Agnostic Design — Core reasoning logic remains unchanged while adapters handle source-specific data normalization and target-specific output formatting.
-
Azure Cloud Native — Deployed on Azure App Services with managed identity authentication, Key Vault secret management, and Azure AI Inference/Content Safety integration.
Getting started
Dev/prod commands, environment variables, and runtime notes.
Run the docs site locally while editing.
Core areas
Shared infrastructure: LLM providers, prompt override system, and cross-domain tools used by both K12 and TAP applications.
AI-powered editing and generation of Emergency Operation Plan (EOP) functional annexes and Quick Reference Guides (QRG), helping K12 schools create and maintain critical safety documentation.
Educational content safety platform performing automated batch tagging and content safety analysis of text/images using Azure AI Content Safety with custom severity thresholds.
Domain-agnostic multi-agent analysis pipeline with temporal analysis, pattern recognition, correlation detection, and explanation generation - serves both K12 and TAP through format adapters.
Production deployment on Azure App Services with managed identity authentication, Azure AI Inference integration, Content Safety API, and Key Vault secret management.
SQL Server schema with workflow execution logs, agent prompt overrides, large payload storage, and domain-specific tables for K12 (check-ins, reports) and TAP (content analysis results).
Common entry points
Browse all docs (with filters)
Use these filters to quickly narrow docs by domain and audience.