Skip to content

test(ai): avoid disk setup for live evaluation policy checks #320

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
ice-black-tea merged 1 commit into from
Nov 15, 2026
Merged

test(ai): avoid disk setup for live evaluation policy checks #320

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 3 additions & 3 deletions tests/ai/test_live_evaluation_tool_effects.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -39,7 +39,7 @@
("read_only", "mcp"), ("live", "tool"),
))
async def test_evaluation_enforces_effect_limit_against_next_step_declarations(
tmp_path: Path, monkeypatch: pytest.MonkeyPatch,
monkeypatch: pytest.MonkeyPatch,
policy: Literal["deny", "read_only", "live"], addition: str,
) -> None:
features = RuntimeFeatures(workspace_factory=None)
Expand Down Expand Up @@ -72,7 +72,7 @@ def respond(messages: list[ModelMessage], info: AgentInfo) -> ModelResponse:
features.model("default", models)
features.tool(read, effect_policy="none")
async with (
Persistence.open("live-effects", storage=PersistenceStorage.sqlite(tmp_path / "state.db"), tenant_id=TENANT) as state,
Persistence.open("live-effects", storage=PersistenceStorage.in_memory(), tenant_id=TENANT) as state,
Runtime.attach(persistence=state, context=CONTEXT, features=features) as runtime,
):
target = Task.from_agent("effects.target", runtime.agents.get())
Expand All @@ -86,7 +86,7 @@ def respond(messages: list[ModelMessage], info: AgentInfo) -> ModelResponse:
)), principal=PRINCIPAL, idempotency_key="dataset")
request = StartEvaluationRequest(EvaluationSpec(
dataset, (CandidateSpec("candidate", task=target.ref),), (rule_scorer(scorer),),
policy=EvaluationPolicy(external_effects=policy, model_fixtures=(models.contract,)),
policy=EvaluationPolicy(allow_volatile=True, external_effects=policy, model_fixtures=(models.contract,)),
), PRINCIPAL, "evaluate")
run = await evaluations.start(request)
await run.wait(timeout_seconds=EVALUATION_COMPLETION_TIMEOUT_SECONDS)
Expand Down
Loading