Velocity: Financial Analysis Tool Review

Overview

Velocity is a financial analysis tool developed by Blotter that aims to provide rapid analysis for stock tickers by aggregating insights from various financial data sources such as SEC filings, news, earnings reports, and institutional data. The tool leverages AI to process large volumes of data quickly, aiming to match the analytical capabilities of top Wall Street analysts.

The tool is designed around a modular architecture where different agents focus on specific types of financial data, working together to create a comprehensive view of a stock's performance. This approach allows for efficient data processing and analysis, significantly reducing the time typically required for such tasks.

Key Features

Velocity's key features include the use of Large Language Models (LLMs) to extract valuable insights from financial data. The tool employs multiple agents, each responsible for a specific aspect of financial data, such as SEC filings, news, earnings reports, and code analysis. These agents work in unison to gather and process information, generating outputs like price targets, bull/bear cases, and risk/reward assessments.

The insights gathered are combined into a cohesive output, saved in both JSON and HTML formats for easy review and further analysis. This modular approach ensures that the tool can be customized and fine-tuned to meet specific needs, making it a versatile solution for financial analysis.

Agent-Based Architecture

Velocity's architecture is built around the concept of agents, each designed to focus on a specific type of financial data. These agents include the SECAgent for SEC filings, NewsAgent for financial news, EarningsAgent for earnings reports, and CodingAgent for automated code-based analysis. The Analyst agent synthesizes the final report, including risk themes and stock price targets.

This agent-based approach allows for a more efficient and comprehensive analysis of financial data, enabling the tool to process large volumes of information quickly and accurately.

Large Language Models (LLMs)

Velocity utilizes Large Language Models (LLMs) to interpret and analyze the gathered insights. These models help in self-reflection and interpretation of the data, generating detailed outputs such as price targets, bull/bear cases, and risk/reward assessments. The use of LLMs enhances the tool's ability to provide deep and meaningful insights from the financial data.

Command-Line Interface

Velocity can be run from the command line, taking in several parameters to configure its behavior. The required parameters include the stock ticker symbol (`--ticker`), OpenAI API key (`--openai_key`), and Financial Modeling Prep (FMP) API key (`--fmp_key`). These keys can also be set as environment variables (`OPENAI_API_KEY` and `FMP_API_KEY`) if not provided via the command line.

An example command to run Velocity is: `python3.9 velocity.py --ticker AAPL --openai_key <YOUR_OPENAI_API_KEY> --fmp_key <YOUR_FMP_API_KEY>`. Note that the tool currently only supports Python 3.9.

Configuration File

The `config.py` file is used to store configurations related to different agent types. This file allows users to customize the types of agents being used, set cache expiration times for insights, and manage data sources for each agent. By editing `config.py`, users can fine-tune the behavior of the agents and adjust the overall analysis flow to better suit their needs.

Technical Requirements

Velocity requires Python 3.9 to run. Users must ensure that this version of Python is available on their system and can be accessed via the command `python3.9`. Additionally, the tool requires API keys from OpenAI and Financial Modeling Prep (FMP) to function, which can be provided either as command-line arguments or set as environment variables.

Output Formats

The insights generated by Velocity are saved in both JSON and HTML formats. This dual format allows for easy review and further analysis, catering to different user preferences and needs. The HTML format provides a more visually accessible report, while the JSON format is ideal for programmatic processing and integration with other tools.

License

Velocity is released under the Apache-2.0 license. This license allows for free use, modification, and distribution of the software, provided that the terms of the license are adhered to. The Apache-2.0 license also includes a warranty disclaimer, stating that the software is provided 'as-is' without any warranties or guarantees.

Support and Maintenance

As an open-source project, Velocity does not come with formal support or maintenance guarantees. However, the community and contributors may provide assistance and updates. Users are encouraged to report issues and contribute to the project on its GitHub repository to help improve the tool over time.