Agentic Ticker Research
Performs multi-step, reasoning-based stock research by analyzing news, prices, and market sentiment through an agentic AI backend powered by Gemini and FastAPI.
Acerca de
This tool functions as an autonomous financial analyst, processing queries like 'Find the latest news about AAPL and link it with its price changes in the last 30 days.' It systematically collects recent price trends, fetches and summarizes news headlines from sources like Yahoo Finance, links them to stock movements, and finally provides a reasoning-based analysis. Built with a verified, modular Agentic Cognitive Architecture, it leverages Gemini for prompt validation, exposes tools via the MCP framework, and uses Pydantic-typed cognitive layers to ensure explicit reasoning, structured output, and robust fallback mechanisms, delivering a comprehensive market overview.
Características Principales
- Abstracted MCP tools with Pydantic schemas for data validation
- Transcript memory to store user sessions, turns, and preferences for context
- 0 GitHub stars
- Modular 4-layer Agentic Cognitive Architecture (Perception, Memory, Decision, Action)
- Resilient LLM call handling with overload error management and prompt truncation
- Gemini-driven multi-step reasoning with deterministic function calls and final answers
Casos de Uso
- Automated multi-step stock research and analysis for specific tickers
- Correlating recent financial news with historical stock price changes
- Generating reasoning-based summaries of market events and their impact on stocks
