Every AI finance tool today treats investing as one-shot: ask a question, get an answer, move on. But real investing is Bayesian — you start with a thesis, new data arrives daily, and you update your conviction accordingly. It's an iterative process that unfolds over weeks and mo
How to install langalpha?
ginlix-ai/langalpha
$git clone https://github.com/ginlix-ai/langalphaInstalls into the current project.
›Prefer a prompt? Paste this to your agent
Install langalpha by running `git clone https://github.com/ginlix-ai/langalpha`, then use it for the current task and follow its documentation at https://github.com/ginlix-ai/langalpha.
README.md
| 1 | <p align="center"> |
| 2 | <img src="web/public/logo_words.png" alt="LangAlpha" height="120" /> |
| 3 | <br> |
| 4 | <strong>A vibe investing agent harness</strong> |
| 5 | <br> |
| 6 | LangAlpha is built to help interpret financial markets and support investment decisions. |
| 7 | <br><br> |
| 8 | <img src="https://img.shields.io/badge/python-3.12+-blue.svg" alt="Python 3.12+" /> |
| 9 | <a href="https://github.com/langchain-ai/langchain"><img src="https://img.shields.io/badge/LangChain-1c3c3c?logo=langchain&logoColor=white" alt="LangChain" /></a> |
| 10 | <img src="https://img.shields.io/badge/license-Apache%202.0-green.svg" alt="License" /> |
| 11 | </p> |
| 12 | |
| 13 | <p align="center"> |
| 14 | <strong>English</strong> ｜ <a href="docs/README.zh-CN.md">简体中文</a> ｜ <a href="docs/README.ja-JP.md">日本語</a> |
| 15 | </p> |
| 16 | |
| 17 | <p align="center"> |
| 18 | <a href="#getting-started">Getting Started</a> • |
| 19 | <a href="docs/api/README.md">API Docs</a> • |
| 20 | <a href="src/ptc_agent/">Agent Core</a> • |
| 21 | <a href="src/server/">Backend</a> • |
| 22 | <a href="web/">Web</a> • |
| 23 | <a href="libs/ptc-cli/">TUI</a> • |
| 24 | <a href="skills/">Skills</a> • |
| 25 | <a href="mcp_servers/">MCP</a> |
| 26 | </p> |
| 27 | |
| 28 | <p align="center"> |
| 29 | <video src="https://github.com/user-attachments/assets/56ec23b5-e9af-46ab-8505-66a7dff822a4" autoplay loop muted playsinline width="900"></video> |
| 30 | </p> |
| 31 | <p align="center"><em>Pin a curated news brief from the dashboard, kick off idea generation, and dispatch parallel subagents to screen the market — then get five long/short pair-trade ideas in an inline interactive dashboard, calibrated to your book.</em></p> |
| 32 | |
| 33 | ## Why LangAlpha |
| 34 | |
| 35 | Every AI finance tool today treats investing as one-shot: ask a question, get an answer, move on. But real investing is Bayesian — you start with a thesis, new data arrives daily, and you update your conviction accordingly. It's an iterative process that unfolds over weeks and months: refining theses, revisiting positions, layering new analysis on top of old. No single prompt captures that. |
| 36 | |
| 37 | ### *From vibe coding to vibe investing* |
| 38 | |
| 39 | Inspired by software engineering: a codebase persists, and every commit builds on what came before. Code agent harnesses like Claude Code and OpenCode succeeded by building agents that embrace this pattern, exploring existing context and building on prior work. LangAlpha brings that same insight: give the agent a persistent workspace, and research naturally compounds. |
| 40 | |
| 41 | In practice, you create a workspace per research goal ("Q2 rebalance", "data center demand deep dive", "energy sector rotation"). The agent interviews you about your goals and style, produces its first deliverable, and saves everything to the workspace filesystem. Come back tomorrow and your files, threads, and accumulated research are still there. |
| 42 | |
| 43 | ## Features Highlights |
| 44 | |
| 45 | - **Progressive Tool Discovery** — Any MCP tools loaded as summary in context and full documentation dumped into the workspace, allowing the agent to discover and use tools truly on demand. Also supports binding json tools with skills and only expose to agent when skill is activated. |
| 46 | - **Programmatic Tool Calling (PTC)** — The agent writes and executes Python to process financial data from mcp servers instead of pouring raw data into the LLM context window, enabling complex multi-step analysis while dramatically reducing token waste. |
| 47 | - **Financial data ecosystem** — Multi-tier provider hierarchy with native tools for quick lookups and MCP servers for bulk data processing, charting, and multi-year analysis in sandboxes. |
| 48 | - **Persistent workspaces** — Each workspace maps to a dedicated sandbox with structured directories and a workspace notes file (`agent.md`) that compounds research across sessions and threads. A separate long-term memory store (`.agents/user/memory/`, `.agents/workspace/memory/`) persists durable user preferences and cross-sandbox knowledge, and a user-managed memo store (`.agents/user/memo/`) lets you upload PDFs and markdown research notes that the agent can read on demand. |
| 49 | - **Skills for Financial Research** — Pre-built |
Related
6 picksType
- financetoolkit200+ transparent financial metrics calculated from raw statements, not third-party endpoints.Jul 2026
- mcp-serverAn MCP server for interacting with the Financial Datasets stock market API.Jun 2025
- mcp-serverPython MCP server for local interactions with the QuantConnect API.May 2026
- alpaca-mcp-serverAlpaca Trading API integration for Model Context Protocol (MCP). V2 generates 60+ tools directly from Alpaca's OpenAPI specs using FastMCP.Jul 2026
- finopsmcpSee where your cloud and AI bills go, and spend less. Local-first, in your terminal or editor.Jul 2026
- xero-mcp-serverMCP server implementation for Xero integrationJun 2026
