Open Source Agent Simulation SDK
Ship Reliable AI Agents
A modern testing framework that transforms AI agents into production-grade systems.
$
Star on GitHubpip install tenro@tenro.simulate
def test_cancel_order():
# Simulate different scenarios
llm.simulate(
Provider.ANTHROPIC,
responses=[
ToolCall(cancel_order, order_id="123"),
"Done! Order #123 has been cancelled.",
],
)
tool.simulate(cancel_order, result="cancelled")
# Run your agent
support_agent.execute("Cancel my order #123")
# Verify tool calls and agent's final output
tool.verify(cancel_order, order_id="123")
agent.verify(support_agent, result="Done! Order #123 has been cancelled.")How it works
Link. Simulate. Verify.
Battle test your agents. Full control, zero API costs.
01
Link
@link_agent | @link_toolDecorate your agents, tools, and LLM calls.
02
Simulate
llm.simulate(...)Control external responses. No API calls during tests.
03
Verify
agent.verify(...)Assert calls happened with expected inputs and outputs.
Simulation engine for agents
Test without burning through tokens. Simulate LLM responses, verify agent behavior, and run your test suite in milliseconds.
0
API calls in tests
100%
reproducible
ms
not seconds
Integrations
Works with your stack
Compatible with major LLM providers and agent frameworks.
LLM Providers
OpenAI
Anthropic
Google Gemini
Agent Frameworks
LangChain
LangGraph
LlamaIndex
CrewAI
AutoGen
PydanticAI
Custom
Stay in the loop
Get early access & updates
Be the first to know about new features. No spam.