agentic-vnstock

A lightweight, zero-bloat library to fetch Vietnam stock market data, perfectly designed for AI Agents, LLMs, and Data Analysts.

Features (Phase 1)

• Fetch historical stock prices (OHLCV) for any ticker on HOSE, HNX, and UPCOM.
• Output clean, ready-to-use Pandas DataFrames or CSV files.
• Zero dependencies on bloated third-party charting libraries; uses only requests and pandas.
• Built-in CLI for instant data extraction without writing Python code.

Installation

pip install agentic-vnstock

Quick CLI Usage

You can use the built-in command-line interface to quickly download stock data to a CSV file. The basic syntax is:

agentic-vnstock stock <TICKER> <START_DATE> <END_DATE>

Example: Fetch FPT stock data for the first half of 2026:

agentic-vnstock stock FPT 2026-01-01 2026-06-30

This will create a file named FPT_data.csv in your current directory containing Date, Open, High, Low, Close, and Volume.

(Note: If you are on Windows and get a "command not found" error because your Python Scripts folder isn't in your PATH, you can always run it via Python directly:)

python -m agentic_vnstock.cli stock FPT 2026-01-01 2026-06-30

Python Usage

Use it directly in your Python scripts, Jupyter Notebooks, or AI Agent tools.

from agentic_vnstock import AgenticVNStock

client = AgenticVNStock()

# Fetch historical price data for HPG from Jan 1, 2026 to Apr 21, 2026
df = client.get_historical_price("HPG", "2026-01-01", "2026-04-21")
print(df.head())

Architecture Roadmap

• Phase 1 (Current): Blazing fast data fetching from TCBS APIs (client.py) and easy CLI export.
• Phase 2 (Upcoming): Built-in AI Agents (agents/) and Tools (tools/) optimized for LLM integrations (LangChain, LlamaIndex, OpenAI function calling).