skilltest-pytest 0.3.0__tar.gz → 0.4.0__tar.gz

{skilltest_pytest-0.3.0 → skilltest_pytest-0.4.0}/PKG-INFO

@@ -1,12 +1,12 @@
 Metadata-Version: 2.4
 Name: skilltest-pytest
-Version: 0.3.0
+Version: 0.4.0
 Summary: pytest integration for skilltest: auto-collect *.skilltest.yaml cases as pytest tests, built on skilltest-sdk.
 Author: Nick DeRobertis
 License-Expression: MIT
 Requires-Python: >=3.12
 Requires-Dist: pytest>=8
-Requires-Dist: skilltest-sdk==0.3.0
+Requires-Dist: skilltest-sdk==0.4.0
 Description-Content-Type: text/markdown
 
 # skilltest-pytest

{skilltest_pytest-0.3.0 → skilltest_pytest-0.4.0}/pyproject.toml

@@ -1,13 +1,13 @@
 [project]
 name = "skilltest-pytest"
-version = "0.3.0"
+version = "0.4.0"
 description = "pytest integration for skilltest: auto-collect *.skilltest.yaml cases as pytest tests, built on skilltest-sdk."
 readme = "README.md"
 requires-python = ">=3.12"
 license = "MIT"
 authors = [{ name = "Nick DeRobertis" }]
 dependencies = [
-    "skilltest-sdk==0.3.0",
+    "skilltest-sdk==0.4.0",
     "pytest>=8",
 ]
 

{skilltest_pytest-0.3.0 → skilltest_pytest-0.4.0}/skilltest_pytest/__init__.py

@@ -25,10 +25,13 @@ from skilltest_sdk import (
     Message,
     NumericDetail,
     Report,
+    SkillStream,
     SkilltestError,
     SkilltestProviderError,
     SkilltestUsageError,
+    StreamEvent,
     Summary,
+    ToolEvent,
     Transcript,
     Usage,
     ValidationFinding,
@@ -38,6 +41,8 @@ from skilltest_sdk import (
     failed_evals,
     failed_runs,
     run_skill,
+    stream_skill,
+    tool_calls,
     validate_skill,
 )
 
@@ -52,11 +57,14 @@ __all__ = [
     "Message",
     "NumericDetail",
     "Report",
+    "SkillStream",
     "SkilltestError",
     "SkilltestFailure",
     "SkilltestProviderError",
     "SkilltestUsageError",
+    "StreamEvent",
     "Summary",
+    "ToolEvent",
     "Transcript",
     "Usage",
     "ValidationFinding",
@@ -66,5 +74,7 @@ __all__ = [
     "failed_evals",
     "failed_runs",
     "run_skill",
+    "stream_skill",
+    "tool_calls",
     "validate_skill",
 ]

{skilltest_pytest-0.3.0 → skilltest_pytest-0.4.0}/tests/test_plugin.py

@@ -8,11 +8,12 @@ whose eval fails — can be asserted on too.
 
 from __future__ import annotations
 
+import asyncio
 from pathlib import Path
 
 import pytest
 
-from skilltest_pytest import describe_failures, run_skill
+from skilltest_pytest import describe_failures, run_skill, stream_skill, tool_calls
 
 SKILL_MD = """\
 ---
@@ -40,6 +41,20 @@ def test_sdk_api_is_reexported_and_works(cases: Path) -> None:
     assert report.passed, describe_failures(report)
 
 
+def test_tool_events_and_streaming_reexported_and_work(cases: Path) -> None:
+    # The tool-event and streaming surfaces are re-exported from the plugin too.
+    report = run_skill(cases / "tool_events.yaml")
+    assert [c.name for c in tool_calls(report.runs[0].transcript)] == ["edit_file", "bash"]
+
+    async def go() -> list[str | None]:
+        names: list[str | None] = []
+        async for ev in stream_skill(cases / "tool_events.yaml"):
+            names.append(ev.event.name)
+        return names
+
+    assert asyncio.run(go()) == ["edit_file", "bash"]
+
+
 def test_collected_case_passes(pytester: pytest.Pytester) -> None:
     write_skill(pytester.path)
     pytester.makefile(

{skilltest_pytest-0.3.0 → skilltest_pytest-0.4.0}/uv.lock

@@ -189,7 +189,7 @@ wheels = [
 
 [[package]]
 name = "skilltest-pytest"
-version = "0.3.0"
+version = "0.4.0"
 source = { editable = "." }
 dependencies = [
     { name = "pytest" },
@@ -216,7 +216,7 @@ dev = [
 
 [[package]]
 name = "skilltest-sdk"
-version = "0.3.0"
+version = "0.4.0"
 source = { editable = "../../sdks/python" }
 dependencies = [
     { name = "pydantic" },