AI Stock Market News Analyzer with Trading Signals

AI-Powered Market Intelligence System Setup Guide

What This Agent Does

This workflow automates financial market intelligence gathering by continuously monitoring multiple news sources, analyzing sentiment, and generating actionable trading insights. Every 15 minutes, the system aggregates breaking news from Bloomberg, CNBC, and Reuters, applies AI-powered sentiment analysis, and delivers alerts through your preferred communication channels while maintaining a comprehensive audit trail.

Key benefits and time savings:

• Eliminates manual news monitoring – Automatically tracks three major financial news sources 24/7
• Real-time decision support – Receive AI-analyzed market signals within minutes of breaking news
• Intelligent filtering – Only alerts you to actionable opportunities, reducing noise and alert fatigue
• Comprehensive documentation – Maintains searchable records in Airtable and Google Sheets for compliance and analysis
• Multi-channel delivery – Ensures you never miss critical alerts via Slack and Telegram

Target use cases:

• Active traders and portfolio managers monitoring market movements
• Financial analysts tracking sector-specific news and sentiment shifts
• Risk management teams identifying emerging market threats
• Investment research teams building data-driven intelligence reports

Who Is It For

This workflow is ideal for financial professionals who need real-time market intelligence without constant manual monitoring. Whether you're a day trader, portfolio manager, financial analyst, or investment strategist, this system transforms raw market data into actionable insights. You'll benefit most if you:

• Monitor multiple news sources daily
• Make time-sensitive investment decisions
• Want AI-powered sentiment analysis without building custom tools
• Need documented decision trails for compliance
• Prefer automated alerts over manual checking

Required Integrations

RSS Feed Reader

Why it's needed: Connects to Bloomberg Markets, CNBC Breaking News, and Reuters Business feeds to continuously pull the latest financial news without manual intervention.

Setup steps:

1. No API key required – RSS is a standard web protocol
2. Verify that your news sources offer public RSS feeds (most major financial outlets do)
3. In TaskAGI, the RSS integration is typically pre-configured
4. Test feed connectivity by running a single feed node before full deployment

How to obtain credentials:

• RSS feeds are publicly available; no authentication needed
• Feed URLs are typically found on news websites under "Subscribe" or "RSS" sections
• Example URLs:
  • Bloomberg: https://www.bloomberg.com/feed/podcast/etf-report.xml
  • CNBC: https://www.cnbc.com/id/100003114/device/rss/rss.html
  • Reuters: https://www.reuters.com/finance

Configuration in TaskAGI:

• Each RSS node requires only the feed URL
• Set polling frequency to match your 15-minute trigger interval
• Enable error handling to gracefully skip unavailable feeds

OpenAI API

Why it's needed: Powers two critical AI functions – sentiment analysis of news articles and generation of trading strategy recommendations based on market signals.

Setup steps:

1. Visit platform.openai.com and sign in or create an account
2. Navigate to API keys in the left sidebar
3. Click Create new secret key and copy the generated key immediately
4. Store the key securely (you won't see it again)
5. In TaskAGI, go to Integrations → OpenAI
6. Paste your API key in the authentication field
7. Verify the connection with a test call

How to obtain API keys:

• Requires an OpenAI account with billing enabled
• Free trial credits are available but limited
• Production use requires a paid account (pay-as-you-go pricing)
• Monitor usage in your OpenAI dashboard to control costs

Configuration in TaskAGI:

• Model: gpt-4o (specified in workflow for optimal financial analysis)
• Temperature: 0.7 (balances creativity with consistency for analysis)
• Max tokens: 1500 (sufficient for detailed sentiment analysis and strategy generation)
• Store your API key as a secure environment variable, never hardcode it

Slack

Why it's needed: Delivers real-time alerts to your team when actionable trading signals are identified, ensuring immediate awareness of market opportunities.

Setup steps:

1. Open your Slack workspace and go to Settings & administration → Manage apps
2. Search for "TaskAGI" or create a custom app
3. Click Create New App → From scratch
4. Name it "Market Intelligence Bot" and select your workspace
5. Under OAuth & Permissions, add these scopes: chat:write, channels:read
6. Click Install to Workspace and authorize
7. Copy the Bot User OAuth Token (starts with xoxb-)
8. In TaskAGI, navigate to Integrations → Slack
9. Paste the token and select your target channel (e.g., #trading-alerts)

How to obtain credentials:

• Requires Slack workspace admin access
• OAuth tokens are generated automatically during app installation
• Keep tokens confidential – treat as passwords

Configuration in TaskAGI:

• Channel: #trading-alerts (or your preferred channel)
• Message format: Include timestamp, sentiment score, and action recommendation
• Enable message threading to keep related alerts organized

Telegram

Why it's needed: Provides a secondary alert channel for critical signals, ensuring you receive notifications even if Slack is unavailable, with mobile push notifications.

Setup steps:

1. Open Telegram and search for BotFather
2. Send /newbot and follow prompts to create your bot
3. BotFather will provide a Bot Token (save this)
4. Create a Telegram group or channel for alerts
5. Add your bot to the group and make it an administrator
6. Get your Chat ID by sending a message to the group and checking bot logs
7. In TaskAGI, go to Integrations → Telegram
8. Enter the Bot Token and Chat ID

How to obtain credentials:

• Bot Token: Provided by BotFather (format: 123456:ABC-DEF1234ghIkl-zyx57W2v1u123ew11)
• Chat ID: Numeric identifier for your group/channel (format: -1001234567890)

Configuration in TaskAGI:

• Parse mode: HTML (for formatted messages)
• Include emoji indicators for sentiment (🟢 bullish, 🔴 bearish, 🟡 neutral)
• Test with a sample message before enabling production alerts

Airtable

Why it's needed: Creates a searchable, structured database of all market signals, sentiment analysis results, and trading strategies for compliance, backtesting, and historical analysis.

Setup steps:

1. Sign in to airtable.com or create an account
2. Create a new base named "Market Intelligence"
3. Create a table with columns: Date, Source, Headline, Sentiment Score, Signal Type, Strategy, Status
4. Go to Account → API (or Developer Hub)
5. Generate a Personal Access Token with data.records:read and data.records:write scopes
6. In TaskAGI, navigate to Integrations → Airtable
7. Paste your token and select your base and table

How to obtain credentials:

• Personal Access Token: Generated in Airtable account settings
• Base ID: Found in your base URL (format: appXXXXXXXXXXXXXX)
• Table ID: Found in Airtable API documentation for your table

Configuration in TaskAGI:

• Create records with all signal metadata for audit trails
• Update records with strategy recommendations and execution status
• Set up views for filtering by sentiment, source, or date range

Google Sheets

Why it's needed: Maintains a lightweight, shareable log of all market signals in a familiar spreadsheet format for team collaboration and quick reference.

Setup steps:

1. Create a new Google Sheet titled "Market Intelligence Log"
2. Add headers: Timestamp, Source, Headline, Sentiment, Action, Notes
3. Go to Share and grant edit access to your team
4. Copy the Sheet URL from your browser
5. In TaskAGI, go to Integrations → Google Sheets
6. Authenticate with your Google account
7. Paste the sheet URL in the configuration

How to obtain credentials:

• Google account with Google Sheets access
• OAuth authentication (TaskAGI handles this automatically)
• Sheet URL: Full URL from your browser address bar

Configuration in TaskAGI:

• Sheet URL: Paste the complete URL (note: the workflow shows sheet_url: null – you must populate this)
• Append mode: Add new rows without overwriting existing data
• Format dates consistently (ISO 8601: YYYY-MM-DD HH:MM:SS)

Configuration Steps

Node-by-Node Configuration Guidance

1. Schedule Trigger (Every 15 Minutes)

• Set interval to 900 seconds (15 minutes)
• This determines how frequently the entire workflow executes
• Adjust based on your monitoring needs (more frequent = higher API costs)

2. RSS Feed Nodes (Bloomberg, CNBC, Reuters)

• Bloomberg Markets Feed: https://www.bloomberg.com/feed/podcast/etf-report.xml
• CNBC Breaking News: https://www.cnbc.com/id/100003114/device/rss/rss.html
• Reuters Business: https://www.reuters.com/finance
• Each node outputs: title, description, pubDate, link

3. Aggregate News Sources (Merge Node)

• Combines all three RSS feeds into a single data stream
• Output: Array of news items with standardized fields
• Removes duplicates if the same story appears across sources

4. Filter & Extract Signals (Function Node)

• Purpose: Identify market-moving keywords (earnings, merger, bankruptcy, FDA approval, etc.)
• Logic: Filter articles containing financial signal keywords
• Output: Filtered articles with relevance scores

5. AI Sentiment Analysis (OpenAI Node)

• Prompt: "You are a financial analyst AI. Analyze the following market news and provide: 1) Sentiment score (-1 to +1), 2) Key entities mentioned, 3) Market impact assessment."
• Model: gpt-4o
• Input: Filtered news articles
• Output: Structured sentiment analysis with scores and reasoning

6. Fetch Stock Data (Function Node)

• Purpose: Retrieve current stock prices for mentioned companies
• Logic: Extract ticker symbols from articles and fetch real-time data
• Output: Current price, 52-week range, volume, market cap

7. Synthesize Intelligence Report (Function Node)

• Purpose: Combine sentiment analysis with stock data into actionable insights
• Logic: Correlate sentiment with price movements and volatility
• Output: Structured report with signal strength and confidence levels

8. Check If Actionable (Conditional Node)

• Condition: sentiment_score > 0.6 OR signal_strength > 0.7
• True path: Proceed to alerts and strategy generation
• False path: Log to database without alerts (no action needed)

9. Send Slack Alert (Slack Node)

• Message format: Include headline, sentiment score, recommended action
• Channel: #trading-alerts
• Trigger: Only when condition is true

10. Send Telegram Alert (Telegram Node)

• Message format: Concise version with emoji indicators
• Trigger: Same as Slack (parallel execution)

11. Log to Airtable (Airtable Node)

• Record fields: All analysis results, timestamps, sources
• Trigger: Always (both actionable and non-actionable signals)

12. Log to Google Sheets (Google Sheets Node)

• Row data: Timestamp, source, headline, sentiment, action
• Trigger: Always
• Important: Update the sheet_url parameter with your actual sheet URL

13. Generate Trading Strategy (OpenAI Node)

• Prompt: "You are a portfolio strategist. Based on the trading signal and market data provided, generate: 1) Recommended action (buy/sell/hold), 2) Entry/exit points, 3) Risk management rules, 4) Position sizing guidance."
• Model: gpt-4o
• Trigger: Only when actionable condition is true

14. Update Airtable with Strategy (Airtable Node)

• Action: Update the record created in step 11 with strategy details
• Fields: Recommended action, entry point, exit point, risk level

Testing Your Agent

Step 1: Pre-Deployment Testing

1. Verify all integrations:

   • Test each API connection individually
   • Confirm RSS feeds are accessible and returning data
   • Validate OpenAI API key and model availability
   • Test Slack and Telegram message delivery
   • Confirm Airtable and Google Sheets write permissions

2. Run a single workflow execution:

   • Click Test in the TaskAGI interface
   • Monitor the execution log for errors
   • Verify data flows correctly through each node

Step 2: Data Flow Validation

1. Check RSS aggregation:

   • Confirm all three feeds return articles
   • Verify the merge node combines them correctly
   • Expected: 10-50 articles per 15-minute cycle

2. Validate AI analysis:

   • Review sentiment scores (should range from -1 to +1)
   • Confirm stock data is fetched for mentioned companies
   • Check that the intelligence report synthesizes both data sources

3. Test conditional logic:

   • Manually trigger with high-sentiment articles
   • Verify alerts are sent (Slack + Telegram)
   • Manually trigger with low-sentiment articles
   • Verify no alerts are sent, but logging still occurs

Step 3: Output Verification

1. Slack alerts:

   • Check that messages appear in #trading-alerts
   • Verify formatting is readable and includes all key data
   • Confirm timestamps are accurate

2. Telegram alerts:

   • Receive test message in your Telegram group
   • Verify emoji indicators display correctly
   • Check mobile push notification delivery

3. Airtable records:

   • Open your "Market Intelligence" base
   • Confirm new records appear with all fields populated
   • Verify strategy updates are applied to existing records

4. Google Sheets log:

   • Open your "Market Intelligence Log" sheet
   • Confirm new rows are appended (not overwriting)
   • Check that all columns are populated correctly

Step 4: Expected Results and Success Indicators

Successful execution should show:

• ✅ 3 RSS feeds returning articles every 15 minutes
• ✅ Sentiment scores calculated for all filtered articles
• ✅ Slack and Telegram alerts sent for high-confidence signals
• ✅ Complete audit trail in Airtable with timestamps
• ✅ Readable log in Google Sheets for team reference
• ✅ Trading strategies generated for actionable signals
• ✅ Zero errors in execution logs

Common issues and solutions:

• No articles returned: Verify RSS feed URLs are current (news sites occasionally change them)
• API rate limits: Increase the 15-minute interval or upgrade your OpenAI plan
• Missing Google Sheets data: Confirm sheet_url parameter is populated with your actual sheet URL
• Alerts not sending: Verify Slack/Telegram tokens are valid and channels are accessible

Congratulations! Your AI-powered market intelligence system is now live. You'll receive real-time alerts on actionable market signals while maintaining a comprehensive audit trail for compliance and analysis. Start with the 15-minute interval and adjust based on alert volume and your trading frequency.