github-featured/openai--openai-agents-python

Files

T

History

Mohammed Siddiq 09f0ed45e7 fix: enforce the documented strict JSON-compatible contract for #3486 (#3657 )

2026-06-18 08:26:47 +00:00

..

fix: expose sandbox error retryability (#3581 )

2026-06-05 10:53:06 -07:00

Fix: Correct streaming order for ReasoningItem and RawResponsesStreamEvent events (#1869 )

2025-10-08 15:16:06 -07:00

feat: add SDK-only custom data for tool outputs (#3486 )

2026-06-18 12:38:28 +09:00

fix: #3268 fix OpenAI Conversations reasoning persistence (#3352 )

2026-05-12 07:20:47 +09:00

Runtime handling updates (#3451 )

2026-05-18 16:53:20 -07:00

test: add unit tests for _openai_retry helpers (#3544 )

2026-06-04 07:38:26 +00:00

feat: add pre-approval tool input guardrails (#3487 )

2026-06-18 12:38:15 +09:00

fix: expose sandbox error retryability (#3581 )

2026-06-05 10:53:06 -07:00

fix: make tracing shutdown best-effort on process exit (#3343 )

2026-05-11 15:19:32 +09:00

fix: guard None text in ItemHelpers.extract_last_content (#3394 )

2026-05-15 15:40:30 +09:00

fix: #3363 honor short custom voice splitter chunks (#3364 )

2026-05-16 10:04:23 +09:00

__init__.py

Initial commit

2025-03-11 09:42:28 -07:00

_fake_workspace_paths.py

feat: support sandbox extra path grants (#2920 )

2026-04-17 15:26:51 -07:00

conftest.py

test: improve coverage and organize test layout (#3085 )

2026-05-02 15:40:54 +09:00

fake_model.py

Sandbox Agents (#2889 )

2026-04-15 10:00:40 -07:00

README.md

perf: speed up make tests with xdist and deterministic waits (#2563 )

2026-03-01 12:42:55 +09:00

test_agent_as_tool.py

fix: handle cancelled single function tools as tool failures (#2762 )

2026-03-25 14:17:20 +09:00

test_agent_clone_shallow_copy.py

Fix: Emit tool_called events immediately in streaming runs (#1300 )

2025-08-25 10:48:32 +09:00

test_agent_config.py

chore: drop Python 3.9 support across metadata, CI, and docs (#2474 )

2026-02-13 11:39:12 -08:00

test_agent_hooks.py

fix: #3512 type tool-end hook results as object (#3518 )

2026-05-28 17:57:33 +09:00

test_agent_instructions_signature.py

feat: #636 Add human-in-the-loop (HITL) support (#2230 )

2026-01-29 12:16:28 +09:00

test_agent_llm_hooks.py

fix: #3512 type tool-end hook results as object (#3518 )

2026-05-28 17:57:33 +09:00

test_agent_memory_leak.py

fix: #2008 Fix agent memory leak using weakref (#2014 )

2025-11-18 17:09:50 +09:00

test_agent_prompt.py

fix: invalid model setting when passing prompt to Agent (#1852 )

2025-11-19 13:33:43 +09:00

test_agent_runner_streamed.py

fix: #3459 add opt-in recovery for missing function tools (#3461 )

2026-05-21 10:46:44 +09:00

test_agent_runner_sync.py

Sandbox Agents (#2889 )

2026-04-15 10:00:40 -07:00

test_agent_runner.py

feat: add pre-approval tool input guardrails (#3487 )

2026-06-18 12:38:15 +09:00

test_agent_tool_input.py

docs: sync tool docs and add coverage for internal helper branches (#2485 )

2026-02-13 22:43:09 -08:00

test_agent_tool_state.py

test: improve coverage for tracing and runtime helpers (#2635 )

2026-03-08 02:22:05 +00:00

test_agent_tracing.py

Sandbox Agents (#2889 )

2026-04-15 10:00:40 -07:00

test_agents_logging.py

Add new tools for gpt-5.1 (#2079 )

2025-11-13 09:55:29 -08:00

test_apply_diff_helpers.py

Add CRLF support for apply_diff (#2394 )

2026-02-03 22:06:14 +09:00

test_apply_diff.py

Fix minor typos in provider comments and test name (#2674 )

2026-03-15 14:41:58 +09:00

test_apply_patch_tool.py

fix: #3359 preserve local approval rejection reasons (#3360 )

2026-05-11 22:32:46 +00:00

test_asyncio_progress.py

fix: isolate parallel function tool failures (#2584 )

2026-03-04 05:37:10 +09:00

test_call_model_input_filter_unit.py

fix: #2038 openai==2.7.0 compatibility (#2039 )

2025-11-05 10:23:28 +09:00

test_call_model_input_filter.py

fix: preserve latest session tool outputs and keep orphan calls in non-resume input prep (#2411 )

2026-02-04 18:41:26 +09:00

test_cancel_streaming.py

fix: #2929 surface run-loop exceptions after stream_events() completes (#2931 )

2026-04-18 09:01:57 +09:00

test_computer_action.py

fix: #3512 type tool-end hook results as object (#3518 )

2026-05-28 17:57:33 +09:00

test_computer_tool_lifecycle.py

fix: exclude Computer instances from provider duck-typing (#3249 )

2026-05-09 02:45:31 +09:00

test_config.py

feat: expose Responses WebSocket keepalive options (#3080 )

2026-05-02 09:18:45 +09:00

test_custom_tool.py

fix: #3359 preserve local approval rejection reasons (#3360 )

2026-05-11 22:32:46 +00:00

test_daytona_usaspending_example.py

chore: improve automated example coverage and local service handling (#3297 )

2026-05-09 16:17:49 +09:00

test_debug.py

Fix a bug where the default behavior of loading the debug flag differs from the documentation (#1682 )

2025-09-08 10:05:25 -04:00

test_doc_parsing.py

Initial commit

2025-03-11 09:42:28 -07:00

test_example_workflows.py

Sandbox Agents (#2889 )

2026-04-15 10:00:40 -07:00

test_exception_exports.py

fix(exceptions): export MCPToolCancellationError from top-level package (#3210 )

2026-05-09 01:58:48 +09:00

test_extension_filters.py

fix: filter hosted_tool_call types in remove_all_tools handoff filter (#3386 )

2026-05-15 15:40:04 +09:00

test_extra_headers.py

Fix #1564 Add conversations API support (#1587 )

2025-08-29 10:06:36 -04:00

test_function_schema.py

feat: support Annotated[T, Field(...)] in function schema (#2435 )

2026-02-09 14:25:32 -08:00

test_function_tool_decorator.py

Sandbox Agents (#2889 )

2026-04-15 10:00:40 -07:00

test_function_tool.py

fix: redact invalid JSON payload in ModelBehaviorError data (#3485 )

2026-05-21 23:54:12 +00:00

test_global_hooks.py

fix: #3512 type tool-end hook results as object (#3518 )

2026-05-28 17:57:33 +09:00

test_guardrails.py

fix: #3280 streaming guardrail exception cleanup (#3281 )

2026-05-09 02:16:03 +09:00

test_handoff_history_duplication.py

Sandbox Agents (#2889 )

2026-04-15 10:00:40 -07:00

test_handoff_prompt.py

feat: #636 Add human-in-the-loop (HITL) support (#2230 )

2026-01-29 12:16:28 +09:00

test_handoff_tool.py

fix: suppress handoff whitespace tool-name warnings (#3652 )

2026-06-18 10:29:58 +09:00

test_hitl_error_scenarios.py

fix: skip needs_approval_checker when status already resolved (#3229 )

2026-05-09 02:37:35 +09:00

test_hitl_session_scenario.py

feat: #636 Add human-in-the-loop (HITL) support (#2230 )

2026-01-29 12:16:28 +09:00

test_hitl_utils.py

feat: #636 Add human-in-the-loop (HITL) support (#2230 )

2026-01-29 12:16:28 +09:00

test_items_helpers.py

feat: #2886 add convenience properties (tool_name, call_id) to tool items (#3027 )

2026-04-26 10:32:30 +09:00

test_local_shell_tool.py

feat: add container shell + skills support (#2469 )

2026-02-11 09:56:03 -08:00

test_logprobs.py

Add logprobs to ModelSettings (#971 )

2025-08-11 13:59:44 +09:00

test_max_turns.py

feat: allow disabling max_turns with None (#3132 )

2026-05-06 21:14:27 +09:00

test_mcp_tool_metadata.py

test: add direct unit tests for _mcp_tool_metadata helpers (#3102 )

2026-05-04 18:00:20 +09:00

test_oaiconv_resume_response_id.py

fix(run): preserve last known response_id on conversation resume (#3245 )

2026-05-09 02:42:05 +09:00

test_output_guardrail_cancellation.py

fix: await cancelled output guardrail tasks on tripwire (#3187 )

2026-05-09 11:43:15 +09:00

test_output_tool.py

fix: #3357 output schema names for Literal types (#3358 )

2026-05-14 19:24:35 +09:00

test_pretty_print.py

fix: guard None text in text_message_output and add output guardrail count to RunErrorDetails (#3375 )

2026-05-12 14:53:34 +09:00

test_process_model_response.py

fix: #3459 add opt-in recovery for missing function tools (#3461 )

2026-05-21 10:46:44 +09:00

test_prompt_cache_key.py

feat: add context management model setting (#3128 )

2026-05-05 20:55:16 +09:00

test_repl.py

test: add unit tests for run_demo_loop streaming, EOF, and empty-input paths (#3542 )

2026-06-04 07:32:26 +00:00

test_responses_tracing.py

fix: reject failed responses stream terminals (#3107 )

2026-05-04 22:20:46 +09:00

test_responses.py

fix: #3055 surface model refusals during run resolution (#3057 )

2026-05-01 14:28:13 +09:00

test_result_cast.py

feat: expose immutable agent tool invocation metadata on run results (ref #2575 ) (#2576 )

2026-03-02 14:06:30 +09:00

test_run_config.py

fix: #3459 add opt-in recovery for missing function tools (#3461 )

2026-05-21 10:46:44 +09:00

test_run_context_approvals.py

Runtime handling updates (#3451 )

2026-05-18 16:53:20 -07:00

test_run_context_wrapper.py

feat: #2658 preserve explicit approval rejection messages across resume flows (#2660 )

2026-03-13 14:23:32 +09:00

test_run_error_details.py

Don't cache agent tools during a run (#803 )

2025-06-02 14:49:16 -04:00

test_run_examples_script.py

feat: improve examples auto-run coverage and artifact handling (#3328 )

2026-05-10 14:25:47 +09:00

test_run_hooks.py

fix: #3512 type tool-end hook results as object (#3518 )

2026-05-28 17:57:33 +09:00

test_run_impl_resume_paths.py

fix #2151 shield server-managed handoffs from unsupported history rewrites (#2747 )

2026-04-16 03:17:00 +09:00

test_run_internal_approvals.py

test: improve coverage and organize test layout (#3085 )

2026-05-02 15:40:54 +09:00

test_run_internal_error_handlers.py

docs: sync tool docs and add coverage for internal helper branches (#2485 )

2026-02-13 22:43:09 -08:00

test_run_internal_items.py

fix: drop reasoning items orphaned by dropped tool calls (#3207 )

2026-05-09 02:06:58 +09:00

test_run_state.py

feat: add SDK-only custom data for tool outputs (#3486 )

2026-06-18 12:38:28 +09:00

test_run_step_execution.py

fix: #3512 type tool-end hook results as object (#3518 )

2026-05-28 17:57:33 +09:00

test_run_step_processing.py

Sandbox Agents (#2889 )

2026-04-15 10:00:40 -07:00

test_run.py

feat: #636 Add human-in-the-loop (HITL) support (#2230 )

2026-01-29 12:16:28 +09:00

test_runner_guardrail_resume.py

feat: #636 Add human-in-the-loop (HITL) support (#2230 )

2026-01-29 12:16:28 +09:00

test_server_conversation_tracker.py

fix: #3004 serve HITL resume tool outputs (#3006 )

2026-04-23 10:36:15 +09:00

test_shell_call_serialization.py

fix: reject string-like shell commands (#3092 )

2026-05-04 09:50:59 +09:00

test_shell_tool.py

fix: #3359 preserve local approval rejection reasons (#3360 )

2026-05-11 22:32:46 +00:00

test_soft_cancel.py

fix: #2171 dedupe nested handoff inputs (#2323 )

2026-01-16 20:24:13 +09:00

test_source_compat_constructors.py

feat: add pre-approval tool input guardrails (#3487 )

2026-06-18 12:38:15 +09:00

test_stream_events.py

fix: skip CompactionItem silently in stream queue helper (#3224 )

2026-05-08 16:27:11 +00:00

test_stream_input_guardrail_timing.py

perf: speed up make tests with xdist and deterministic waits (#2563 )

2026-03-01 12:42:55 +09:00

test_streamed_terminal_output_backfill.py

fix: backfill streamed terminal output (#3000 )

2026-04-23 09:32:44 +09:00

test_streaming_logging.py

feat: #636 Add human-in-the-loop (HITL) support (#2230 )

2026-01-29 12:16:28 +09:00

test_streaming_tool_call_arguments.py

Sandbox Agents (#2889 )

2026-04-15 10:00:40 -07:00

test_strict_schema_oneof.py

Sandbox Agents (#2889 )

2026-04-15 10:00:40 -07:00

test_strict_schema.py

fix: #3317 return fresh empty strict schemas (#3318 )

2026-05-10 08:54:25 +09:00

test_tool_choice_reset.py

feat: add Responses API tool search support (#2610 )

2026-03-06 19:23:38 +09:00

test_tool_context.py

fix: make ToolContext hashable to match RunContextWrapper (#3097 )

2026-05-04 09:58:28 +09:00

test_tool_converter.py

feat: add Responses API tool search support (#2610 )

2026-03-06 19:23:38 +09:00

test_tool_custom_data.py

fix: enforce the documented strict JSON-compatible contract for #3486 (#3657 )

2026-06-18 08:26:47 +00:00

test_tool_guardrails.py

test: cover guardrail name fallback to function __name__ (#3073 )

2026-05-02 08:14:47 +09:00

test_tool_identity.py

test: add direct unit tests for _tool_identity helpers (#3101 )

2026-05-04 17:59:43 +09:00

test_tool_metadata.py

fix: show the GA computer alias in tracing (#2641 )

2026-03-09 15:22:25 +09:00

test_tool_origin.py

feat: #2228 persist tool origin metadata in run items (#2654 )

2026-04-16 05:01:04 +09:00

test_tool_output_conversion.py

Make Tool Output dict conversion stricter to improve backward compatibility (#1965 )

2025-10-22 09:37:17 +09:00

test_tool_use_behavior.py

feat: add Responses API tool search support (#2610 )

2026-03-06 19:23:38 +09:00

test_tool_use_tracker.py

Sandbox Agents (#2889 )

2026-04-15 10:00:40 -07:00

test_trace_processor.py

fix: #3354 interrupt tracing retry backoff on shutdown (#3355 )

2026-05-12 07:23:11 +09:00

test_tracing_errors_streamed.py

fix: redact invalid JSON payload in ModelBehaviorError data (#3485 )

2026-05-21 23:54:12 +00:00

test_tracing_errors.py

fix: redact invalid JSON payload in ModelBehaviorError data (#3485 )

2026-05-21 23:54:12 +00:00

test_tracing_provider_safe_debug.py

Harden tracing shutdown logging against closed handlers (#2270 )

2026-01-08 13:41:38 +09:00

test_tracing.py

Sandbox Agents (#2889 )

2026-04-15 10:00:40 -07:00

test_transforms.py

fix: #2951 warn for tool name character replacement (#2953 )

2026-04-19 02:18:10 +00:00

test_usage.py

fix: preserve existing request_usage_entries on Usage.add (#3213 )

2026-05-09 02:07:20 +09:00

test_usaspending_setup_db.py

feat: improve examples auto-run coverage and artifact handling (#3328 )

2026-05-10 14:25:47 +09:00

test_visualization.py

test: cover real Handoff object branch in visualization (#3100 )

2026-05-04 17:58:14 +09:00

testing_processor.py

Sandbox Agents (#2889 )

2026-04-15 10:00:40 -07:00

README.md

Tests

Before running any tests, make sure you have uv installed (and ideally run make sync after).

Running tests

make tests

make tests runs the shard-safe suite in parallel and then runs tests marked serial in a separate serial pass.

Snapshots

We use inline-snapshots for some tests. If your code adds new snapshot tests or breaks existing ones, you can fix/create them. After fixing/creating snapshots, run make tests again to verify the tests pass.

Fixing snapshots

make snapshots-fix

Creating snapshots

make snapshots-create