Korea Investment & Securities (KIS) Open API Sample Codes

The strategy_builder project is organized into several key modules that separate the user interface, the API server, and the core strategy engine:

• backend/: A FastAPI server running on port 8000. It provides API endpoints for authentication, accounts, strategy management, orders, market data, and file handling.
• strategy_core/: The Single Source of Truth (SSoT) for the strategy engine. It includes the StrategyRegistry (using the @register decorator), the executor, and a DSL (Domain Specific Language) layer that converts builder states into Python code via a parser and codegen.
• core/: Low-level modules for signal/action definitions, technical indicators (80+ types), KIS API data fetching, order execution, position management, and WebSocket handling.
• strategy/: Contains the BaseStrategy class and specific strategy implementations (strategy_01 to strategy_10.py).
• frontend/: A Next.js application running on port 3000. It includes the visual Strategy Builder UI, the execution UI, and hooks for interacting with the backend.
• kis_auth.py: The module used for Korea Investment & Securities (KIS) API authentication.

This repository provides sample code for the Korea Investment & Securities (KIS) Open API, designed for both Python developers and LLM-based AI agents. It is organized into two main types of examples:

1. examples_llm/: Optimized for LLMs. It contains granular, single-functionality samples. Each API feature has its own folder containing a minimal execution file ([function_name].py) and a verification/test file (chk_[function_name].py).
2. examples_user/: Optimized for human developers. It provides integrated, category-based samples. Each category (e.g., domestic_stock) contains integrated function files ([category]_functions.py) and execution example files ([category]_examples.py). It also includes WebSocket versions (_ws.py).

Core Components:

• kis_auth.py: Handles authentication (access tokens, WebSocket keys), common API call functions, and switching between real and mock investment environments.
• strategy_builder/: A visual UI engine for designing trading strategies and generating BUY/SELL/HOLD signals.
• backtester/: A QuantConnect Lean-based engine for validating strategies against historical data.
• kis_devlp.yaml: Configuration file for API settings (requires personal credentials).

.
├── strategy_builder/            # 전략 설계 + 시그널 생성 엔진
├── backtester/                  # 백테스팅 엔진 (QuantConnect Lean)
├── docs/
│   └── convention.md            # 코딩 컨벤션 가이드
├── examples_llm/                  # LLM용 샘플 코드
│   ├── kis_auth.py              # 인증 공통 함수
│   └── ...
├── examples_user/                 # user용 실제 사용 예제
│   ├── kis_auth.py              # 인증 공통 함수
│   └── ...
├── kis_devlp.yaml               # API 설정 파일 (개인정보 입력 필요)
└── pyproject.toml               # (uv)프로젝트 의존성 관리

The Strategy Builder and Backtester are designed to work together through a unified .kis.yaml file format, enabling a complete quantitative workflow: Design → Validation → Execution.

Workflow Steps:

1. Strategy Design: Use the Strategy Builder visual UI to configure indicators, conditions, and risk parameters.
2. Export: Export the designed strategy as a .kis.yaml file.
3. Backtest: Import the .kis.yaml file into the Backtester to validate strategy performance using historical data (Lean).
4. Analysis: Analyze results (returns, Sharpe ratio, Max Drawdown) within the Backtester.
5. Optimization: Use Grid or Random Search in the Backtester to find optimal parameters.
6. Live Application: Once validated, bring the strategy back to the Strategy Builder for real-time signal generation and paper/live trading.

Comparison of Capabilities:

You can define trading strategies without writing code by using the .kis.yaml format. This allows for dynamic parameter binding using the $param_name syntax.

YAML Schema Structure

• metadata: Name, description, author, and tags.
• strategy: ID, category, and params (defining default, min, max, type, and description).
• indicators: List of indicators with id, alias, and params (which can reference strategy parameters).
• entry/exit: Logic (e.g., AND, OR) and conditions (indicator, operator, and value).
• risk: Risk management settings like stop_loss.

Example YAML Strategy

version: "1.0"

metadata:
  name: "RSI 과매도 반전"
  description: "RSI가 과매도 영역 진입 후 반등 시 매수"
  author: "user"
  tags: [oscillator, rsi, mean_reversion]

strategy:
  id: rsi_oversold
  category: oscillator

  params:
    period:
      default: 14
      min: 2
      max: 100
      type: int
      description: "RSI 기간"
    oversold:
      default: 30
      min: 0
      max: 50
      type: float
      description: "과매도 기준"

  indicators:
    - id: rsi
      alias: rsi
      params:
        period: $period

  entry:
    logic: AND
    conditions:
      - indicator: rsi
        operator: cross_above
        value: $oversold

  exit:
    logic: OR
    conditions:
      - indicator: rsi
        operator: cross_below
        value: 70

risk:
  stop_loss:
    enabled: true
    percent: 5.0

The .kis.yaml format is the shared Single Source of Truth (SSoT) used by both Strategy Builder and Backtester. It defines the strategy ID, parameters, indicators, entry/exit logic, and risk management.

Supported Operators

Example .kis.yaml Strategy

version: "1.0"

metadata:
  name: "SMA 골든/데드 크로스"
  description: "단기 SMA가 장기 SMA 돌파 시 매매"
  author: "KIS"
  tags: [trend, sma, crossover]

strategy:
  id: sma_crossover
  category: trend

  params:
    fast_period:
      default: 20
      min: 2
      max: 100
      type: int
    slow_period:
      default: 50
      min: 2
      max: 200
      type: int

  indicators:
    - id: sma
      alias: sma_fast
      params:
        period: $fast_period
    - id: sma
      alias: sma_slow
      params:
        period: $slow_period

  entry:
    logic: AND
    conditions:
      - indicator: sma_fast
        operator: cross_above
        compare_to: sma_slow

  exit:
    logic: OR
    conditions:
      - indicator: sma_fast
        operator: cross_below
        compare_to: sma_slow

risk:
  stop_loss:
    enabled: true
    percent: 5.0
  take_profit:
    enabled: true
    percent: 10.0