Deep research agent for investment & strategy teams. Synthesizes financial statements, earnings calls & market data into structured investment memos. Features dual research modes, persistent memory via Qdrant, contradiction detection, and scenario analysis — built on OpenAI Agents SDK.
Stars
2
Forks
0
Watchers
2
Open Issues
0
Overall repository health assessment
No package.json found
This might not be a Node.js project
13
commits
Add initial implementation of finance agent backend, including core modules for agent runtime, API routes, and models. Introduce agents for deep analysis, document handling, and memory management. Set up caching and Qdrant service for data management. Add .gitignore for environment files and Python artifacts.
6d0def9View on GitHubIntegrate session management into AppShell and Sidebar components. Implement session storage functionality with hooks for adding and restoring sessions. Update Sidebar to display recent sessions and handle session restoration on user interaction. Enhance useChat hook to support restoring conversations from saved sessions.
658b580View on GitHubAdd Qdrant status endpoint to API for connectivity and collections verification, enhancing health check capabilities.
bef1311View on GitHubAdd CORS support for production environment by including the new origin for the frontend application hosted on Vercel.
c71dfbbView on GitHubAdd production Docker configuration for frontend service. Introduce Dockerfile and nginx configuration for serving static files. Update README with production deployment instructions and environment variable requirements.
4e52f03View on GitHubUpdate Docker configuration and health checks for backend services. Add documentation exclusion to .dockerignore. Refactor backend structure by removing unused files and updating requirements for improved dependency management. Enhance README with local setup instructions for backend and frontend.
d742413View on GitHubRefactor backend agents for financial analysis by restructuring agent modules under a new `fin_agents` directory. Update API routes to reflect the new module paths. Introduce new agents: Deep, Document, Memory, Quick, and Triage, each with specific functionalities for financial research. Enhance agent instructions and improve document ingestion and retrieval processes for better user experience.
459d9fbView on GitHubEnhance backend functionality by updating agent runtime to support list input for user queries and adding a fallback streaming method. Implement CORS middleware in FastAPI for cross-origin requests. Introduce memory update request model and improve API routes for memory management. Add dotenv support for environment variable loading across various modules. Update frontend API client to handle CORS and improve SSE handling with new event types.
a3c1721View on GitHubAdd Docker support with Dockerfile and docker-compose.yml for backend and frontend services. Create .dockerignore to exclude unnecessary files from the Docker context. Update README.md to reflect project changes and API documentation for chat and analysis endpoints.
ef3f6a7View on GitHubAdd backend structure for finance agent application, including initialization files, agent runtime, API routes, and models. Implemented caching and Qdrant service for data management. Introduced various tools for financial analysis and document handling.
492464dView on GitHubAdd backend agents for financial analysis, including Deep, Memory, Quick, and Triage agents. Implemented .gitignore for environment files and Python artifacts. Established API routes for chat, analysis, memory management, document ingestion, health checks, and cache stats. Integrated Qdrant for document storage and semantic search capabilities.
ceb6bdeView on GitHubAdd initial README.md with project overview, features, architecture, and setup instructions for FinSight
a11b3bdView on GitHub