Analyze the time and space complexity of your run method and what approach would you take to test this implementation?

Time complexity: The run method performs a constant amount of work per step (tool selection, execution, error handling, history trimming). If the step limit is S, the overall time is O(S). Space complexity: The history list is capped at 10 messages, so space is O(1) relative to input size. The tool objects themselves may use additional space, but that is outside the orchestrator’s responsibility.

Testing approach:

1. Unit tests for normal flow – mock tools that return predictable outputs; verify that history contains the expected sequence and that the step limit is respected.
2. Error handling tests – mock a tool that raises an exception; ensure the failure counter increments, the fallback tool is called, and the loop stops after the failure limit.
3. History truncation tests – feed more than 10 messages and confirm that only the last 10 remain.
4. Boundary tests – run with step limit 0, failure limit 0, and empty tool list to ensure graceful handling.
5. Performance tests – measure execution time for a large step limit to confirm linear scaling.

Mocking frameworks (e.g., unittest.mock) allow isolation of the orchestrator from actual tool implementations, making these tests deterministic and fast.