How to Run Unit Tests for Agent Components in TradingAgents-CN

Use pytest -k "agents" after installing dependencies to execute only the agent-specific unit tests, or run pytest to execute the full suite while automatically excluding integration tests.

TradingAgents-CN is an open-source multi-agent trading system that relies on rigorous unit testing to ensure the reliability of its analyst, manager, and risk-debator components. This guide explains how to run unit tests for the agent components using the pytest framework configured in the repository.

Prerequisites and Environment Setup

Before executing tests, you must install the project dependencies and activate the virtual environment.

# Clone the repository

git clone https://github.com/hsliuping/TradingAgents-CN.git
cd TradingAgents-CN

# Create and activate a virtual environment

python -m venv .venv
source .venv/bin/activate  # Windows: .venv\Scripts\activate

# Install dependencies

pip install -r requirements.txt

The requirements.txt file at the repository root defines all necessary packages, including pytest and testing utilities.

Understanding the Test Structure

TradingAgents-CN organizes its test suite under the tests/ directory, with configuration defined in tests/pytest.ini. This configuration restricts test collection to the tests/ folder and automatically skips any test marked with the integration label.

Agent-related unit tests target the core modules located in tradingagents/agents/. Key test files include:

• tests/tradingagents/test_app_cache_toggle.py – Validates agent-level caching mechanisms and data-service wiring.
• tests/unit/tools/analysis/test_indicators_uil.py – Tests technical-indicator utilities consumed by analyst agents.

Running Unit Tests for Agent Components

Run All Unit Tests

To execute the complete suite of unit tests while excluding integration tests (as configured in pytest.ini), run:

pytest

This command collects all tests from the tests/ directory and respects the exclusion rules defined in the configuration file.

Filter for Agent-Specific Tests

To run unit tests for the agent components specifically, use the -k flag to filter by keyword:

pytest -k "agents"

This executes only tests whose node IDs contain the substring "agents", effectively targeting files like test_app_cache_toggle.py and other agent-related modules.

Alternatively, specify the exact test file path:

pytest tests/tradingagents/test_app_cache_toggle.py

Verbose Output and Debugging

For detailed output that shows individual test names and failure details, add the verbose flag:

pytest -v -k "agents"

This format helps identify which specific agent component tests pass or fail during debugging sessions.

Handling Integration Tests

By default, TradingAgents-CN excludes integration tests from the standard pytest run. The pytest.ini configuration automatically skips tests marked with @pytest.mark.integration.

To override this behavior and include integration tests (which may require external services or databases), explicitly target the integration marker:

pytest -m integration

Use this command only when you have configured the necessary environment variables and external dependencies required for full integration testing.

Summary

• TradingAgents-CN uses pytest configured via tests/pytest.ini to manage the test suite.
• Install dependencies with pip install -r requirements.txt before running tests.
• Execute all unit tests with the pytest command, which automatically excludes integration tests.
• Target agent components specifically using pytest -k "agents" or by specifying individual test files like tests/tradingagents/test_app_cache_toggle.py.
• Use pytest -v for verbose output and pytest -m integration to include integration tests when needed.

Frequently Asked Questions

How do I run only the agent caching tests in TradingAgents-CN?

Execute pytest tests/tradingagents/test_app_cache_toggle.py to run only the specific test file that validates agent-level caching and data-service wiring. Alternatively, use pytest -k "cache" to filter for any test containing the word "cache" in its name or path.

What Python version is required to run the unit tests?

The repository requires Python 3.8 or higher, as specified in the requirements.txt and project configuration. Ensure you create your virtual environment with a compatible Python version before installing dependencies and running pytest.

Why are my integration tests being skipped when I run pytest?

The tests/pytest.ini file explicitly configures pytest to skip tests marked with @pytest.mark.integration by default. This design keeps unit test runs fast and isolated from external dependencies. To include integration tests, run pytest -m integration after configuring the necessary environment variables.

Can I run agent tests in parallel to speed up execution?

Yes, install the pytest-xdist package (pip install pytest-xdist) and run pytest -n auto -k "agents" to distribute agent component tests across multiple CPU cores. This significantly reduces execution time when running the full agent test suite.

{
  "mcpServers": {
    "instagit": {
      "command": "npx",
      "args": ["-y", "instagit@latest"]
    }
  }
}