Financial Signals Dashboard

A real-time financial signals agent using Strands Agent SDK with Bright Data MCP integration. The system collects data from financial sources to analyze stocks and generate alpha signals for investment decisions.

Features

• Real-time stock analysis using Bright Data MCP tools
• Interactive Streamlit dashboard with rich visualizations
• AI-powered alpha signals with confidence scores
• Technical analysis with price, moving averages, and RSI indicators
• Position size recommendations
• Risk-reward profile visualization
• Market sentiment analysis with social media metrics
• News source sentiment tracking with visualization
• Support for multiple model providers (AWS Bedrock and Ollama)

Account Setup

Make sure you have an account on brightdata.com (new users get free credit for testing, and pay as you go options are available)

Get your API key from the user settings page https://brightdata.com/cp/setting/users

Setup

1. First, ensure that you have Python 3.10+ installed.

2. Create a virtual environment to install the Strands Agents SDK and its dependencies:

python -m venv .venv

3. Activate the virtual environment:

# macOS / Linux
source .venv/bin/activate

# Windows (CMD)
.venv\Scripts\activate.bat

# Windows (PowerShell)
.venv\Scripts\Activate.ps1

4. Install dependencies:

pip install -r requirements.txt

5. Set your Bright Data API token as an environment variable:

export API_TOKEN="your-api-token-here"

6. [Only required if using Ollama model provider] Install and setup Ollama:

   Option 1: Native Installation

   • Install Ollama by following the instructions at ollama.ai
   • Pull your desired model:

     ollama pull llama3

   • Start the Ollama server:

     ollama serve

   Option 2: Docker Installation

   • Pull the Ollama Docker image:

     docker pull ollama/ollama

   • Run the Ollama container:

     docker run -d -v ollama:/root/.ollama -p 11434:11434 --name ollama ollama/ollama

     Note: Add --gpus=all if you have a GPU and if Docker GPU support is configured.
   • Pull a model using the Docker container:

     docker exec -it ollama ollama pull llama3

   • Verify the Ollama server is running:

     curl http://localhost:11434/api/tags

7. Run the Streamlit app:

streamlit run streamlit_app.py

Model Provider Options

The dashboard supports two model providers:

AWS Bedrock

• Cloud-based model with high performance
• Requires AWS credentials
• Default option for production use

Ollama

• Local model running on your machine
• Requires Ollama to be installed and running
• Supported models:
  • llama3.1:latest (recommended for tool use)
  • llama3:latest
  • llama3.1:latest
  • llama3:8b
  • llama3:70b
  • mistral:latest
  • mixtral:latest

Note on Ollama Tool Support: Standard Ollama models like llama3:latest don't natively support tools and may return errors like registry.ollama.ai/library/llama3:latest does not support tools (status code: 400). We've implemented a workaround using specialized prompting techniques as discussed in this GitHub issue. For best results with tools, use the llama3.1:latest model which has better tool support.

Security Best Practices

Important: Always treat scraped web content as untrusted data. Never use raw scraped content directly in LLM prompts to avoid potential prompt injection risks. Instead:

• Filter and validate all web data before processing
• Use structured data extraction rather than raw text (web_data tools)

Architecture

• Strands Agent SDK: Provides the agent framework for AI-powered analysis
• Bright Data MCP: Handles web scraping and financial data collection
• AWS Bedrock Nova Premier: Powers the AI analysis with advanced language capabilities
• Ollama: Provides local model alternatives for analysis
• Streamlit: Provides the interactive dashboard with real-time updates
• Plotly: Creates interactive and responsive data visualizations

Thread Communication System

The dashboard implements a robust file-based thread communication system:

• Background threads for financial and sentiment analysis
• File-based flags for signaling completion status
• JSON storage for analysis results
• Automatic UI updates when analysis completes
• Error handling with detailed logging and traceback

Dashboard Sections

1. Stock Header: Displays current ticker, price, and date in a visually appealing format

2. Financial Analysis Tab:

   • Signal direction gauge showing BUY/SELL/HOLD recommendation
   • Confidence score gauge with color-coded zones
   • Risk-reward chart comparing potential risk vs. reward
   • Price vs. Moving Averages bar chart
   • RSI gauge showing current RSI value with color-coded zones
   • Position size recommendation display
   • Key factors and risk assessment sections

3. Sentiment Analysis Tab:

   • Market sentiment score gauge with color-coded zones
   • 7-day sentiment trend chart (simulated)
   • Social media sentiment radar chart for Twitter, Reddit, and StockTwits
   • News source sentiment visualization with color-coded bars
   • Detailed news source listings with sentiment analysis

How It Works

1. Enter a stock ticker in the sidebar and click "Analyze Stock"
2. Two background threads are launched for financial and sentiment analysis
3. Strands Agent SDK handles the agentic loop:
   • Manages the agent's reasoning and decision-making process
   • Simplifies tool selection and execution through its React framework
   • Handles the observe-think-act cycle automatically
   • Provides structured output formatting for consistent analysis results
4. The Strands Agent connects to Bright Data MCP and retrieves financial data
5. The AI model analyzes the data and generates structured analysis results
6. Results are saved to files and flags are set to signal completion
7. The UI automatically updates to display the analysis with interactive visualizations
8. Users can explore different aspects of the analysis through the tabbed interface

Technical Details

• Asynchronous processing with proper event loop handling in threads
• Signal extraction using regex patterns to parse structured data from AI responses
• Interactive visualizations powered by Plotly for engaging user experience
• File-based thread communication for reliable UI updates
• Comprehensive error handling with graceful degradation
• Type-safe data handling with proper conversion and validation
• Responsive layout that adapts to different screen sizes
• Model-specific prompting strategies for optimal performance
• Automatic Ollama status checking and model availability verification
• Custom XML-style tool calling format for Ollama models to work around tool support limitations

Investment Disclaimer

The dashboard includes clear disclaimers that:

• The information is for informational purposes only
• The analysis does not constitute investment advice
• Users should consult financial advisors before making investment decisions
• Past performance is not indicative of future results
• Investing involves risk of loss

Future Enhancements

• Real-time price updates
• Portfolio management features
• Historical signal tracking
• Comparative analysis between multiple stocks
• Custom alert settings
• Export functionality for reports
• User accounts for saving analysis history
• Additional model provider integrations