|
| 1 | +--- |
| 2 | +title: "Mastra agents with memory sharing + Trigger.dev task orchestration" |
| 3 | +sidebarTitle: "Mastra agents with memory" |
| 4 | +description: "Multi-agent workflow with persistent memory sharing using Mastra and Trigger.dev for clothing recommendations based on weather data." |
| 5 | +tag: "v4" |
| 6 | +--- |
| 7 | + |
| 8 | +import UpgradeToV4Note from "/snippets/upgrade-to-v4-note.mdx"; |
| 9 | + |
| 10 | +<UpgradeToV4Note /> |
| 11 | + |
| 12 | +## Overview |
| 13 | + |
| 14 | +Enter a city and an activity, and get a clothing recommendation generated for you based on today's weather. |
| 15 | + |
| 16 | + |
| 17 | + |
| 18 | +By combining Mastra's persistent memory system and agent orchestration with Trigger.dev's durable task execution, retries and observability, you get production-ready AI workflows that survive failures, scale automatically, and maintain context across long-running operations. |
| 19 | + |
| 20 | +## Tech stack |
| 21 | + |
| 22 | +- **[Node.js](https://nodejs.org)** runtime environment |
| 23 | +- **[Mastra](https://mastra.ai)** for AI agent orchestration and memory management (Mastra is a Typescript framework for building AI agents, and uses Vercel's AI Agent SDK under the hood.) |
| 24 | +- **[PostgreSQL](https://postgresql.org)** for persistent storage and memory sharing |
| 25 | +- **[Trigger.dev](https://trigger.dev)** for task orchestration, batching, and observability |
| 26 | +- **[OpenAI GPT-4](https://openai.com)** for natural language processing |
| 27 | +- **[Open-Meteo API](https://open-meteo.com)** for weather data (no API key required) |
| 28 | +- **[Zod](https://zod.dev)** for schema validation and type safety |
| 29 | + |
| 30 | +## GitHub repo |
| 31 | + |
| 32 | +<Card |
| 33 | + title="View the Mastra agents with memory repo" |
| 34 | + icon="GitHub" |
| 35 | + href="https://github.com/triggerdotdev/examples/tree/main/mastra-agents" |
| 36 | +> |
| 37 | + Click here to view the full code for this project in our examples repository on GitHub. You can |
| 38 | + fork it and use it as a starting point for your own project. |
| 39 | +</Card> |
| 40 | + |
| 41 | +## Featured patterns |
| 42 | + |
| 43 | +- **[Agent Memory Sharing](https://github.com/triggerdotdev/examples/blob/main/mastra-agents/src/trigger/weather-task.ts)**: Efficient data sharing between agents using Mastra's working memory system |
| 44 | +- **[Task Orchestration](https://github.com/triggerdotdev/examples/blob/main/mastra-agents/src/trigger/weather-task.ts)**: Multi-step workflows with `triggerAndWait` for sequential agent execution |
| 45 | +- **[Centralized Storage](https://github.com/triggerdotdev/examples/blob/main/mastra-agents/src/mastra/index.ts)**: Single PostgreSQL storage instance shared across all agents to prevent connection duplication |
| 46 | +- **[Custom Tools](https://github.com/triggerdotdev/examples/blob/main/mastra-agents/src/mastra/tools/weather-tool.ts)**: External API integration with structured output validation |
| 47 | +- **[Agent Specialization](https://github.com/triggerdotdev/examples/blob/main/mastra-agents/src/mastra/agents/)**: Purpose-built agents with specific roles and instructions |
| 48 | +- **[Schema Optimization](https://github.com/triggerdotdev/examples/blob/main/mastra-agents/src/mastra/schemas/weather-data.ts)**: Lightweight data structures for performance |
| 49 | + |
| 50 | +## Project Structure |
| 51 | + |
| 52 | +``` |
| 53 | +src/ |
| 54 | +├── mastra/ |
| 55 | +│ ├── agents/ |
| 56 | +│ │ ├── weather-analyst.ts # Weather data collection |
| 57 | +│ │ ├── clothing-advisor.ts # Clothing recommendations |
| 58 | +│ ├── tools/ |
| 59 | +│ │ └── weather-tool.ts # Enhanced weather API tool |
| 60 | +│ ├── schemas/ |
| 61 | +│ │ └── weather-data.ts # Weather schema |
| 62 | +│ └── index.ts # Mastra configuration |
| 63 | +├── trigger/ |
| 64 | +│ └── weather-task.ts # Trigger.dev tasks |
| 65 | +``` |
| 66 | + |
| 67 | +## Relevant code |
| 68 | + |
| 69 | +- **[Multi-step task orchestration](https://github.com/triggerdotdev/examples/blob/main/mastra-agents/src/trigger/weather-task.ts)**: Multi-step task orchestration with `triggerAndWait` for sequential agent execution and shared memory context |
| 70 | +- **[Weather analyst agent](https://github.com/triggerdotdev/examples/blob/main/mastra-agents/src/mastra/agents/weather-analyst.ts)**: Specialized agent for weather data collection with external API integration and memory storage |
| 71 | +- **[Clothing advisor agent](https://github.com/triggerdotdev/examples/blob/main/mastra-agents/src/mastra/agents/clothing-advisor.ts)**: Purpose-built agent that reads from working memory and generates natural language responses |
| 72 | +- **[Weather tool](https://github.com/triggerdotdev/examples/blob/main/mastra-agents/src/mastra/tools/weather-tool.ts)**: Custom Mastra tool with Zod validation for external API calls and error handling |
| 73 | +- **[Weather data schema](https://github.com/triggerdotdev/examples/blob/main/mastra-agents/src/mastra/schemas/weather-data.ts)**: Optimized Zod schema for efficient memory storage and type safety |
| 74 | +- **[Mastra configuration](https://github.com/triggerdotdev/examples/blob/main/mastra-agents/src/mastra/index.ts)**: Mastra configuration with PostgreSQL storage and agent registration |
| 75 | + |
| 76 | +## Storage Architecture |
| 77 | + |
| 78 | +This project uses a **centralized PostgreSQL storage** approach where a single database connection is shared across all Mastra agents. This prevents duplicate database connections and ensures efficient memory sharing between the weather analyst and clothing advisor agents. |
| 79 | + |
| 80 | +### Storage Configuration |
| 81 | + |
| 82 | +The storage is configured once in the main Mastra instance (`src/mastra/index.ts`) and automatically inherited by all agent Memory instances. This eliminates the "duplicate database object" warning that can occur with multiple PostgreSQL connections. |
| 83 | + |
| 84 | +The PostgreSQL storage works seamlessly in both local development and serverless environments with any PostgreSQL provider, such as: |
| 85 | + |
| 86 | +- [Local PostgreSQL instance](https://postgresql.org) |
| 87 | +- [Supabase](https://supabase.com) - Serverless PostgreSQL |
| 88 | +- [Neon](https://neon.tech) - Serverless PostgreSQL |
| 89 | +- [Railway](https://railway.app) - Simple PostgreSQL hosting |
| 90 | +- [AWS RDS](https://aws.amazon.com/rds/postgresql/) - Managed PostgreSQL |
| 91 | + |
| 92 | +## Learn More |
| 93 | + |
| 94 | +To learn more about the technologies used in this project, check out the following resources: |
| 95 | + |
| 96 | +- [Mastra docs](https://mastra.ai/en/docs) - learn about AI agent orchestration and memory management |
| 97 | +- [Mastra working memory](https://mastra.ai/en/docs/memory/overview) - learn about efficient data sharing between agents |
0 commit comments