Updated readme and index.js

Author: D-K-P

mastra-agents/.env.example

@@ -0,0 +1,6 @@
+# OpenAI API key for the agent functionality
+OPENAI_API_KEY=<your-openai-api-key>
+
+# Trigger.dev environment variables
+TRIGGER_API_URL=https://api.trigger.dev
+TRIGGER_SECRET_KEY=<your-secret-key>

mastra-agents/README.md

@@ -1,140 +1,27 @@
-# What Should I Wear Today? - Mastra Agents
+# Mastra agents + Trigger.dev real-world example project:
 
-A weather-aware clothing advisor powered by Mastra agents and Trigger.dev. This system provides personalized clothing recommendations based on real-time weather conditions.
+> **ℹ️ Note:** This is a v4 project. If you are using v3 and want to upgrade, please refer to our [v4 upgrade guide](https://trigger.dev/docs/v4-upgrade-guide).
 
-## 🌟 Features
+Enter a city and activity, and get a clothing recommendation generated based on the weather.
 
-- **Weather-Aware Clothing Advice**: Get clothing recommendations based on current temperature, rain chance, and wind speed
-- **Intelligent Memory System**: Uses Mastra's working memory to efficiently share weather data between agents
-- **Activity-Specific Recommendations**: Tailored advice for different activities (walking, running, etc.)
-- **Fast Response**: Optimized for quick responses (~3 seconds)
-- **Simple Input**: Just provide a city and optional activity
-- **Natural Language Output**: Returns human-readable clothing advice paragraph
+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.
 
-## 🏗️ Architecture
+## Tech stack
 
-### Agents
+- [Node.js](https://nodejs.org) runtime environment
+- [Mastra](https://mastra.ai) for AI agent orchestration and memory management
+- [Trigger.dev](https://trigger.dev) for task orchestration, batching, and observability
+- [OpenAI GPT-4](https://openai.com) for natural language processing
+- [Open-Meteo API](https://open-meteo.com) for weather data (no API key required)
+- [Zod](https://zod.dev) for schema validation and type safety
 
-- **Weather Analyst** (`weather-analyst.ts`) - Gets weather data and stores essentials in memory
-- **Clothing Advisor** (`clothing-advisor.ts`) - Reads weather data from memory and provides clothing recommendations
+## Featured patterns
 
-### Tasks
-
-- **`weather-data`** - Gets weather data and stores temperature, rain chance, and wind speed in memory
-- **`clothing-advice`** - Reads weather data from memory and generates clothing recommendations
-- **`what-should-i-wear-today`** - Main task that orchestrates both subtasks with shared memory
-
-### Memory System
-
-- **Simplified Schema** (`weather-data.ts`) - Lightweight weather data structure (4 fields only)
-- **Working Memory** - Stores essential weather data between task executions
-- **Thread-Scoped** - Data shared within same execution thread
-
-## 🚀 Getting Started
-
-### Prerequisites
-
-- Node.js 18+
-- OpenAI API key
-- Trigger.dev account
-
-### Installation
-
-1. **Clone and install dependencies:**
-
-```bash
-git clone <repo-url>
-cd mastra-agents
-npm install
-```
-
-2. **Set up environment variables:**
-
-```bash
-cp .env.example .env
-# Add your OpenAI API key to .env
-OPENAI_API_KEY=your_openai_api_key_here
-```
-
-3. **Install Trigger.dev CLI:**
-
-```bash
-npm install -g @trigger.dev/cli@v4-beta
-```
-
-4. **Start the development server:**
-
-```bash
-npx trigger.dev@v4-beta dev
-```
-
-## 🎯 Usage
-
-### Main Task: What Should I Wear Today?
-
-Get clothing recommendations based on weather:
-
-```bash
-npx trigger.dev@latest dev --trigger what-should-i-wear-today --payload '{"city": "London", "activity": "walking"}'
-```
-
-**Payload:**
-
-- `city` (required): City name (e.g., "London", "New York", "Tokyo")
-- `activity` (optional): Activity type (defaults to "walking")
-
-**How it works:**
-
-1. Weather analyst fetches current weather data and stores temperature, rain chance, and wind speed in memory
-2. Clothing advisor reads from memory and generates clothing recommendations
-3. No duplicate API calls - efficient memory-based task chaining
-
-### Individual Tasks
-
-**Weather Data Only:**
-
-```bash
-npx trigger.dev@latest dev --trigger weather-data --payload '{"city": "Paris"}'
-```
-
-**Clothing Advice Only:**
-
-```bash
-npx trigger.dev@latest dev --trigger clothing-advice --payload '{"city": "Tokyo", "activity": "running", "threadId": "run_abc123"}'
-```
-
-### Sample Output
-
-The main task returns a simple paragraph with clothing advice:
-
-```
-"For walking in London you should wear a light t-shirt or short-sleeved shirt with comfortable pants or shorts, as the temperature is warm at 25°C. With only a 3% chance of rain, you won't need any rain protection, and the light wind at 8 km/h means you can stick to breathable, comfortable clothing for your walk."
-```
-
-## 🧠 Memory System
-
-The system uses Mastra's working memory to share simplified weather data between agents:
-
-```typescript
-// Weather data stored in memory
-{
-  location: "London",
-  temperature: 24.8,      // Current temperature in Celsius
-  rainChance: 3,          // Rain chance percentage (0-100)
-  windSpeed: 8.3          // Wind speed in km/h
-}
-```
-
-This simplified schema focuses on the essential data needed for clothing decisions, making the system faster and more efficient.
-
-## 🛠️ Technical Stack
-
-- **Mastra**: Agent orchestration and memory management
-- **Trigger.dev v4**: Task execution and monitoring
-- **OpenAI GPT-4**: Natural language processing
-- **Open-Meteo API**: Weather data (no API key required)
-- **LibSQL**: Local database for memory storage
-- **Zod**: Type-safe schema validation
+- **[Agent Memory Sharing](src/trigger/weather-task.ts)**: Efficient data sharing between agents using Mastra's working memory system
+- **[Task Orchestration](src/trigger/weather-task.ts)**: Multi-step workflows with `triggerAndWait` for sequential agent execution
+- **[Custom Tools](src/mastra/tools/weather-tool.ts)**: External API integration with structured output validation
+- **[Agent Specialization](src/mastra/agents/)**: Purpose-built agents with specific roles and instructions
+- **[Schema Optimization](src/mastra/schemas/weather-data.ts)**: Lightweight data structures for performance
 
 ## 📁 Project Structure
 
@@ -144,92 +31,61 @@ src/
 │   ├── agents/
 │   │   ├── weather-analyst.ts    # Weather data collection
 │   │   ├── clothing-advisor.ts   # Clothing recommendations
-│   │   └── day-planner.ts        # (Legacy) Activity planning
 │   ├── tools/
 │   │   └── weather-tool.ts       # Enhanced weather API tool
 │   ├── schemas/
-│   │   └── weather-data.ts       # Simplified weather schema
+│   │   └── weather-data.ts       # Weather schema
 │   └── index.ts                  # Mastra configuration
 ├── trigger/
 │   └── weather-task.ts           # Trigger.dev tasks
 └── test-weather-agent.ts         # Local testing
 ```
 
-## 🎨 Customization
+## Relevant code
 
-### Adding New Agents
+- [src/trigger/weather-task.ts](src/trigger/weather-task.ts) - Multi-step task orchestration with `triggerAndWait` for sequential agent execution and shared memory context
+- [src/mastra/agents/weather-analyst.ts](src/mastra/agents/weather-analyst.ts) - Specialized agent for weather data collection with external API integration and memory storage
+- [src/mastra/agents/clothing-advisor.ts](src/mastra/agents/clothing-advisor.ts) - Purpose-built agent that reads from working memory and generates natural language responses
+- [src/mastra/tools/weather-tool.ts](src/mastra/tools/weather-tool.ts) - Custom Mastra tool with Zod validation for external API calls and error handling
+- [src/mastra/schemas/weather-data.ts](src/mastra/schemas/weather-data.ts) - Optimized Zod schema for efficient memory storage and type safety
+- [src/mastra/index.ts](src/mastra/index.ts) - Mastra configuration with LibSQL storage and agent registration
 
-1. Create agent file in `src/mastra/agents/`
-2. Add memory configuration if needed
-3. Register in `src/mastra/index.ts`
+## Getting started
 
-### Modifying Weather Schema
+1. After cloning the repo, run `npm install` to install the dependencies.
+2. Set up your environment variables (see `.env.example`)
+3. If you haven't already, sign up for a free Trigger.dev account [here](https://cloud.trigger.dev/login) and create a new project.
+4. Copy the project ref from the Trigger.dev dashboard and add it to the `trigger.config.ts` file.
+5. In your terminal, run the Trigger.dev dev CLI command with `npx trigger.dev@latest dev`.
 
-Update `src/mastra/schemas/weather-data.ts` to change the weather data structure. The current schema is optimized for speed with just 4 fields.
+Now you should be able to visit your Trigger.dev dashboard and test any of the agent tasks with the example payloads provided in each task file.
 
-### Adding New Activities
+## Testing locally
 
-The system supports different activities. You can extend the clothing advisor agent to provide more specific recommendations for activities like:
+Use the Trigger.dev dashboard to test each task:
 
-- Running
-- Cycling
-- Hiking
-- Formal events
-- Outdoor work
+### Example payload
 
-### Adding New Tasks
-
-Create new tasks in `src/trigger/weather-task.ts` using the existing agents.
-
-## 📊 Features Comparison
-
-| Feature              | Before (Day Planner)   | After (Clothing Advisor) |
-| -------------------- | ---------------------- | ------------------------ |
-| Weather API Calls    | 2+ per execution       | 1 per execution          |
-| Data Schema          | 40+ fields complex     | 4 fields simplified      |
-| Response Time        | 20+ seconds            | ~3 seconds               |
-| Output Format        | Complex JSON structure | Simple text paragraph    |
-| Memory Usage         | Full weather data      | Essential data only      |
-| Agent Specialization | Generic agents         | Purpose-built agents     |
-
-## 🔄 Task Flow
-
-```mermaid
-graph TD
-    A[What Should I Wear Today] --> B[Weather Data Task]
-    B --> C[Weather Analyst Agent]
-    C --> D[Weather Tool - API Call]
-    D --> E[Store in Memory:<br/>temp, rain, wind]
-    E --> F[Clothing Advice Task]
-    F --> G[Clothing Advisor Agent]
-    G --> H[Read from Memory]
-    H --> I[Generate Clothing Advice]
-    I --> J[Return Advice Paragraph]
+```json
+{ "city": "New York", "activity": "walking" }
 ```
 
-## 📈 Performance Benefits
-
-- **Single API call** - Memory system eliminates duplicate weather requests
-- **~3 second responses** - Optimized for speed with simplified data schema
-- **Efficient memory usage** - Only 4 essential fields stored (vs 40+ in full schema)
-- **Better reliability** - Memory persistence across task failures
-- **Cleaner architecture** - Specialized agents with clear responsibilities
+## Deployment
 
-## 🤝 Contributing
+This project uses LibSQL as a local database for development, but **LibSQL doesn't work in serverless environments**. For production deployment, you'll need to switch to a serverless-compatible storage option:
 
-1. Fork the repository
-2. Create a feature branch
-3. Make your changes
-4. Add tests if applicable
-5. Submit a pull request
+- **Turso** (LibSQL-compatible): Set `TURSO_DATABASE_URL` and `TURSO_AUTH_TOKEN` in your environment variables. Create an account and database [here](https://turso.tech/signup).
+- **PostgreSQL** (Supabase): Set `DATABASE_URL` in your environment variables. Create an account and database [here](https://supabase.com/dashboard/sign-in).
+- **No persistence**: Remove storage from mastra config entirely.
 
-## 📄 License
+Update `src/mastra/index.ts` with your chosen storage provider before deploying.
 
-MIT License - see LICENSE file for details
+## Learn More
 
-## 🙏 Acknowledgments
+To learn more about the technologies used in this project, check out the following resources:
 
-- [Mastra](https://mastra.ai) for the agent framework
-- [Trigger.dev](https://trigger.dev) for task orchestration
-- [Open-Meteo](https://open-meteo.com) for weather data
-- [OpenAI](https://openai.com) for language models
+- [Mastra docs](https://docs.mastra.ai) - learn about AI agent orchestration and memory management
+- [Mastra working memory](https://docs.mastra.ai/memory/working-memory) - learn about efficient data sharing between agents
+- [Trigger.dev docs](https://trigger.dev/docs) - learn about Trigger.dev and its features
+- [Trigger.dev task orchestration](https://trigger.dev/docs/triggering#triggering-from-a-task) - learn about sequential task execution with `triggerAndWait`
+- [Multi-agent workflow patterns](https://docs.mastra.ai/agents/multi-agent-workflows) - advanced agent collaboration examples

mastra-agents/src/mastra/index.ts

@@ -3,13 +3,27 @@ import { LibSQLStore } from "@mastra/libsql";
 import { weatherAnalyst } from "./agents/weather-analyst";
 import { clothingAdvisorAgent } from "./agents/clothing-advisor";
 
-export const mastra: Mastra = new Mastra({
-  storage: new LibSQLStore({
+// Storage configuration - for production, switch to serverless-compatible storage
+// See README.md deployment section for options (Turso, PostgreSQL, etc.)
+function createStorage() {
+  // Turso (serverless LibSQL)
+  if (process.env.TURSO_DATABASE_URL) {
+    return new LibSQLStore({
+      url: process.env.TURSO_DATABASE_URL,
+      authToken: process.env.TURSO_AUTH_TOKEN,
+    });
+  }
+
+  // Local development (won't work in serverless)
+  return new LibSQLStore({
     url: "file:./mastra.db",
-  }),
+  });
+}
+
+export const mastra: Mastra = new Mastra({
+  storage: createStorage(),
   agents: {
     weatherAnalyst,
-
     clothingAdvisorAgent,
   },
 });

mastra-agents/trigger.config.ts

@@ -1,7 +1,7 @@
 import { defineConfig } from "@trigger.dev/sdk/v3";
 
 export default defineConfig({
-  project: "proj_xjminyoyogxbrfipkrkt",
+  project: "proj_xjminyoyogxbrfipkrkt", // replace with your project ref
   runtime: "node",
   logLevel: "log",
   // The max compute seconds a task is allowed to run. If the task run exceeds this duration, it will be stopped.