Add Handbook.md benchmark tasks

docker/packages/google_mail/pyproject.toml

@@ -0,0 +1,45 @@
+[build-system]
+build-backend = "uv_build"
+requires = [ "uv-build>=0.11,<0.12" ]
+
+[project]
+name = "google-mail"
+version = "0.1.0"
+description = "Mock email MCP server for RL environment training"
+readme = "README.md"
+requires-python = ">=3.13"
+classifiers = [
+  "Programming Language :: Python :: 3 :: Only",
+  "Programming Language :: Python :: 3.13",
+  "Programming Language :: Python :: 3.14",
+]
+dependencies = [
+  "fastmcp>=3,<4",
+  "pydantic[email]>=2.12.5",
+]
+scripts.google-mail = "google_mail:main"
+
+[dependency-groups]
+dev = [
+  "mypy>=1.19.1",
+  "pytest>=9.0.2",
+  "pytest-asyncio>=1.3",
+  "ruff>=0.14.14",
+]
+
+[tool.ruff]
+target-version = "py313"
+line-length = 120
+lint.select = [ "B", "E", "F", "I", "PLW", "SIM", "UP" ]
+lint.ignore = [ "E501", "PLW0603" ]
+lint.isort.known-first-party = [ "google_mail" ]
+
+[tool.mypy]
+python_version = "3.13"
+strict = true
+warn_return_any = true
+warn_unused_configs = true
+
+[tool.pytest]
+ini_options.testpaths = [ "tests" ]
+ini_options.asyncio_mode = "auto"