agentscope-agent-test-harness

Create and run lightweight tests for AgentScope agents without hitting external APIs. Use this whenever the user wants to validate agent wiring, tool binding, or prompt formatting before live model calls.

This skill helps you test AgentScope agents with mocked models and offline assertions.

What this skill does

• Creates a test harness for an existing agent module
• Replaces the live model with a fake one that returns canned responses
• Asserts tool calls, message formatting, and agent state transitions
• Reports pass/fail in markdown

Usage

Run the harness generator from this skill folder:

bash

python3 scripts/generate_harness.py --agent /path/to/agent_module.py --output /path/to/test_harness.py

Then run:

bash

python3 -m pytest /path/to/test_harness.py

Safety

• No live API calls are made unless the agent explicitly uses a real model.
• Tests are offline by default.