openai-agents-python/tests/mcp
Rohan Mehta d4c7a23e1d
Don't cache agent tools during a run (#803)
### Summary:
Towards #767. We were caching the list of tools for an agent, so if you
did `agent.tools.append(...)` from a tool call, the next call to the
model wouldn't include the new tool. THis is a bug.

### Test Plan:
Unit tests. Note that now MCP tools are listed each time the agent runs
(users can still cache the `list_tools` however).
2025-06-02 14:49:16 -04:00
..
__init__.py [2/n] Add MCP support to Runner 2025-03-24 15:08:02 -04:00
conftest.py [2/n] Add MCP support to Runner 2025-03-24 15:08:02 -04:00
helpers.py [5/n] MCP tracing 2025-03-25 19:28:48 -04:00
test_caching.py [2/n] Add MCP support to Runner 2025-03-24 15:08:02 -04:00
test_connect_disconnect.py [2/n] Add MCP support to Runner 2025-03-24 15:08:02 -04:00
test_mcp_tracing.py Don't cache agent tools during a run (#803) 2025-06-02 14:49:16 -04:00
test_mcp_util.py Ensure MCP works when inputSchema.properties is missing (#454) 2025-04-07 18:38:36 -04:00
test_runner_calls_mcp.py [2/n] Add MCP support to Runner 2025-03-24 15:08:02 -04:00
test_server_errors.py Prevent MCP ClientSession hang (#580) 2025-04-24 12:12:46 -04:00