AI Financial Agent
https://github.com/virattt/ai-financial-agent- Category
- AI Agents
- Rank
- No. 1079Tools index
Previous survey · No. 1065 ·
- Pricing
- Open Source
- Type
- AGENT
- Builder
- virattt
- GitHub
- 2.0k stars
- Date
About
AI agent for investment research — pulls financial filings, runs analyses, and produces investment memos using LLMs.
What it does
A self-hostable browser chat for exploring public-company data. Users ask financial questions, choose an OpenAI model, and receive concise answers alongside generated views for prices, fundamentals, documents, code, and suggestions.
Why it's ranked here
It offers a substantial starting point for financial research experiments, with authentication, persistent conversations, model selection, market-data access, and editable documents already connected. Its proof-of-concept status and explicit educational-only scope keep it from serving as an investment system.
What's good
The project covers more than chat. It stores conversations and votes, supports public or private threads, tracks document versions, highlights suggested edits, and exposes a broad set of statement, valuation, efficiency, liquidity, growth, and per-share filters.
Tradeoffs
Running it requires OpenAI, Financial Datasets, and PostgreSQL credentials, with optional LangSmith tracing. Free market data is limited to five named companies. The maintainers disclaim warranties, financial-loss liability, and suitability for real trading or investment decisions.
How to use it well
Best for developers prototyping a financial research assistant or studying tool-driven chat interfaces. Use it to test questions, data displays, persistent research threads, and editable outputs. It does not replace professional advice, portfolio execution, brokerage connectivity, or production risk controls.
Technical notes+
package.json defines a private Next.js 15 canary application using React 18, TypeScript, Vercel AI SDK, Drizzle ORM, PostgreSQL, Auth.js, ProseMirror, Recharts, and pnpm scripts; its build script runs lib/db/migrate before compiling Next.js. lib/ai/index.ts wraps OpenAI chat models with custom middleware, while lib/ai/models.ts provides four selectable models and defaults to GPT-4o. lib/db/schema.ts models users, chats, messages, votes, versioned documents, and suggestions. lib/db/api-keys.ts reads provider keys from environment variables or browser localStorage. lib/api/stock-filters.ts enumerates supported financial statement, valuation, growth, liquidity, efficiency, and per-share metrics.
Observed
- Primary language
- TypeScript, with React and Next.js
- Install surface
- Private pnpm application installed from source with pnpm install
- Interface
- Browser-based chat application with generative financial displays
- Data storage
- PostgreSQL through Drizzle ORM
- Model provider
- OpenAI through the Vercel AI SDK
- Deployment
- Local Next.js development server or Vercel deployment
Read from README.md, package.json, lib/utils.ts, lib/ai/index.ts, lib/ai/models.ts, lib/db/schema.ts, lib/ai/prompts.ts, lib/db/migrate.ts, lib/db/queries.ts, lib/db/api-keys.ts, lib/editor/diff.js, lib/editor/config.ts, lib/api/stock-filters.ts, lib/editor/functions.tsx, lib/types/data-stream.ts.
What it can do
Pull financial filings
Company ticker or name → Raw financial filing documents
Analyze financial data
Financial filings and statements → Financial metrics and ratios
Generate investment memos
Company research data and analysis results → Structured investment recommendation documents
Research company fundamentals
Company identifier → Business overview and financial performance summary
Extract key financial metrics
Financial statements and filings → Revenue, profit, debt, and growth metrics
Perform comparative analysis
Multiple company financial data → Peer comparison reports and benchmarks
Tags
Tech Stack
Media
Comments (0)
No comments yet
Editorially curated, with community endorsements as a secondary signal. Corrections welcome.