Operator-ready prompt for reuse, tuning, and workspace runs.
This item is set up for developers who want to inspect the original language, fork it into Workspace, and adapt the evidence model without losing the source prompt structure.
Implementation handoffs, eval setup, and prompt tuning where you need the original structure intact.
Inspect first, copy once, then fork into Workspace when you want variants, notes, and model settings attached to the same run.
Swap domain facts, examples, and any hard-coded entities for your own context.
Tighten the evidence or verification requirement if this is headed toward production.
Decide which failure mode you want to evaluate first before you branch the prompt.
This prompt already carries implementation detail, tool context, and a final-output instruction. Keep that structure intact when you tune it, or your comparison runs get noisy fast.
Open this prompt inside Workspace when you want a live iteration loop.
Copy for quick reuse, or run it in Workspace to keep prompt variants, model settings, and prompt-history changes in one place.
Structured source with 100 active lines to adapt.
Already linked to a challenge workflow.
Sign in to keep private prompt variations.
Prompt content
Original prompt text with formatting preserved for inspection and clean copy.
Initialize a Mastra AI agent named `MarketMonitorAgent`. Define tools that allow the agent to:
1. Fetch simulated daily active user (DAU) data for specified platforms.
2. Store and retrieve trend data from Pinecone.
3. Generate a formatted report or summary.
The agent should be configured to use `Mistral Large 2` via the Hugging Face Inference API for its primary reasoning.
```typescript
import { createAgent } from '@mastra-ai/core';
import { Configuration, OpenAIApi } from 'openai'; // Using OpenAI's API client for Pinecone embeddings
import { Pinecone } from '@pinecone-database/pinecone';
// --- Assume these are your API keys and configs ---
const HUGGING_FACE_API_KEY = process.env.HUGGING_FACE_API_KEY;
const PINECONE_API_KEY = process.env.PINECONE_API_KEY;
const PINECONE_ENVIRONMENT = process.env.PINECONE_ENVIRONMENT;
const PINECONE_INDEX_NAME = 'market-trends';
// Initialize Pinecone client
const pinecone = new Pinecone({
environment: PINECONE_ENVIRONMENT,
apiKey: PINECONE_API_KEY,
});
const index = pinecone.Index(PINECONE_INDEX_NAME);
// Simulated DAU data fetcher
const fetchDauData = async (platform: string) => {
console.log(`Fetching DAU data for ${platform}...`);
// In a real scenario, this would hit an external API
const simulatedData = {
'Threads': { 'current': 141.5, 'previous': 138.2, 'unit': 'M DAUs' },
'X': { 'current': 125.0, 'previous': 126.1, 'unit': 'M DAUs' },
'Instagram': { 'current': 600.0, 'previous': 598.5, 'unit': 'M DAUs' }
};
return simulatedData[platform] || null;
};
// Function to store a trend in Pinecone
const storeTrendInPinecone = async (platform: string, trendData: any) => {
console.log(`Storing trend for ${platform} in Pinecone...`);
// Generate embeddings for the trend description using an embedding model
// For simplicity, let's just store metadata for now
const embedding = [0.1, 0.2, 0.3]; // Replace with actual embedding generation
await index.upsert({
vectors: [{
id: `${platform}-${Date.now()}`,
values: embedding,
metadata: { platform, ...trendData, timestamp: new Date().toISOString() },
}],
});
return { status: 'success' };
};
// Function to search for trends in Pinecone
const searchTrendsInPinecone = async (query: string, topK: number = 3) => {
console.log(`Searching Pinecone for: ${query}`);
const queryEmbedding = [0.1, 0.2, 0.3]; // Replace with actual embedding generation for query
const queryResult = await index.query({
vector: queryEmbedding,
topK,
includeMetadata: true,
});
return queryResult.matches.map(match => match.metadata);
};
const MarketMonitorAgent = createAgent({
id: 'MarketMonitorAgent',
model: {
type: 'huggingface',
modelId: 'mistralai/Mistral-Large-2',
apiKey: HUGGING_FACE_API_KEY,
endpointUrl: 'https://api-inference.huggingface.co/models/mistralai/Mistral-Large-2' // Adjust if needed
},
tools: [
{
name: 'fetchDauData',
description: 'Fetches simulated daily active user (DAU) data for a given social media platform.',
inputSchema: { type: 'object', properties: { platform: { type: 'string' } }, required: ['platform'] },
handler: async (args) => fetchDauData(args.platform),
},
{
name: 'storeTrendInPinecone',
description: 'Stores identified market trend data in Pinecone for historical context.',
inputSchema: { type: 'object', properties: { platform: { type: 'string' }, trendData: { type: 'object' } }, required: ['platform', 'trendData'] },
handler: async (args) => storeTrendInPinecone(args.platform, args.trendData),
},
{
name: 'searchTrendsInPinecone',
description: 'Searches for historical market trends in Pinecone based on a query.',
inputSchema: { type: 'object', properties: { query: { type: 'string' }, topK: { type: 'number' } }, required: ['query'] },
handler: async (args) => searchTrendsInPinecone(args.query, args.topK),
}
],
// ... other Mastra config like memory
});
export default MarketMonitorAgent;
// Example of how to run the agent (in a separate file/script):
/*
import MarketMonitorAgent from './MarketMonitorAgent';
async function runAnalysis() {
const result = await MarketMonitorAgent.run({
prompt: 'Analyze the latest DAU data for Threads and identify any significant trends.',
});
console.log('Agent finished:', result);
}
runAnalysis();
*/
```Adaptation plan
Keep the source stable, then branch your edits in a predictable order so the next prompt run is easier to evaluate.
Preserve the role framing, objective, and reporting structure so comparison runs stay coherent.
Swap in your own domain constraints, anomaly thresholds, and examples before you branch variants.
Check whether the prompt asks for the right evidence, confidence signal, and escalation path.
Copy once for a pristine source snapshot, then move the prompt into Workspace when you want variants, run history, and side-by-side tuning without losing the original.
Prompt diagnostics
Quick signals for how structured this prompt already is and where adaptation work is likely to happen first.
This prompt already mixes executable detail with instructions, so the safest path is to tune examples and interfaces before you rewrite the overall scaffold.
Real-time Market Trend Agent
Develop a real-time market trend analysis agent using Mastra AI, designed to monitor specific metrics (e.g., user engagement on social platforms, e-commerce order volumes) and identify significant shifts. The agent will leverage Mistral Large 2 via the Hugging Face Inference API for advanced pattern recognition and sentiment analysis. It will utilize Pinecone as a vector store to maintain a historical context of trends and associated data points, preventing information overload by focusing on novel insights. A Bito AI-like interface component will allow users to interact with the agent for on-demand reports and trend explanations, requiring efficient, scalable processing to deliver timely insights.
Use the challenge page to recover the original task boundaries before you tune the prompt. That keeps your variants grounded in the same evaluation target instead of drifting into a different problem.