Python: Add tau2 benchmark integration with comprehensive testing and documentation (#817)

* first commit to tau2-bench

* tau2-bench agent

* tau2 agent

* add condition

* checkpoint

* bug fix

* add tests

* fix tests

* add comments

* add comments

* minor fix

* fix

* batch test script

* .

* init.bak -> init.py

* fix mypy

* update readme

* fix env

* remove temp files

* setup tests

* fix gaia tasks

* fix tau2 tests

* fix coverage

* fix default version

* update cookiecutter template

---------

Co-authored-by: Eric Zhu <ekzhu@users.noreply.github.com>

python/packages/lab/gaia/agent_framework_lab_gaia/__init__.py

@@ -4,9 +4,16 @@
 GAIA benchmark module for Agent Framework.
 """
 
+import importlib.metadata
+
 from ._types import Evaluation, Evaluator, Prediction, Task, TaskResult, TaskRunner
 from .gaia import GAIA, GAIATelemetryConfig, gaia_scorer, viewer_main
 
+try:
+    __version__ = importlib.metadata.version(__name__)
+except importlib.metadata.PackageNotFoundError:
+    __version__ = "0.0.0"  # Fallback for development mode
+
 __all__ = [
     "GAIA",
     "GAIATelemetryConfig",
@@ -19,5 +26,3 @@ __all__ = [
     "TaskRunner",
     "Evaluator",
 ]
-
-__version__ = "0.1.0b1"

python/packages/lab/gaia/pyproject.toml

@@ -72,7 +72,9 @@ packages = ["agent_framework_lab_gaia"]
 [tool.poe]
 executor.type = "uv"
 include = "../../../shared_tasks.toml"
+
 [tool.poe.tasks]
+test = "pytest --cov=agent_framework_lab_gaia --cov-report=term-missing:skip-covered tests"
 
 [tool.pytest.ini_options]
 testpaths = ["tests"]