feat: finalize production cleanup with structured agent responses and project governance

tests/test_cli.py

@@ -0,0 +1,68 @@
+from __future__ import annotations
+
+import json
+from pathlib import Path
+from typing import Any
+
+import src.cli as cli_module
+from src.config import AppConfig
+from src.models import AgentResult, RunArtifacts, UsageSummary
+
+
+def test_cli_emits_structured_return_and_data(monkeypatch: Any, capsys, tmp_path: Path) -> None:
+    config = AppConfig(
+        openai_api_key="test_key",
+        screenjob_token="test_token",
+        disable_ui=False,
+        default_model="gpt-5.4-mini",
+        safety_model="gpt-5.4-mini",
+        host="127.0.0.1",
+        port=8787,
+        runs_dir=tmp_path / "runs",
+        db_path=tmp_path / "screenjob.db",
+    )
+    config.runs_dir.mkdir(parents=True, exist_ok=True)
+
+    def fake_load_app_config(_: Path) -> AppConfig:
+        return config
+
+    def fake_assess_task_safety(*_args, **_kwargs):
+        return True, "safe", {"safe": True}
+
+    def fake_run_job(*_args, **_kwargs):
+        result = AgentResult(
+            completed=True,
+            result="Done",
+            return_message="Task completed successfully",
+            data="file1.txt\nfile2.txt",
+            steps=3,
+            started_at=10.0,
+            ended_at=12.5,
+            usage=UsageSummary(total_tokens=123),
+            error=None,
+            cancelled=False,
+        )
+        artifacts = RunArtifacts(
+            run_id="20260527_000001",
+            root_dir=config.runs_dir / "run_20260527_000001",
+            logs_dir=config.runs_dir / "run_20260527_000001" / "logs",
+            shots_dir=config.runs_dir / "run_20260527_000001" / "shots",
+            enhance_dir=config.runs_dir / "run_20260527_000001" / "enhance",
+            log_file=config.runs_dir / "run_20260527_000001" / "screenjob.log",
+        )
+        return result, artifacts
+
+    monkeypatch.setattr(cli_module, "load_app_config", fake_load_app_config)
+    monkeypatch.setattr(cli_module, "assess_task_safety", fake_assess_task_safety)
+    monkeypatch.setattr(cli_module, "run_job", fake_run_job)
+    monkeypatch.setattr(cli_module, "create_openai_client", lambda *_args, **_kwargs: object())
+
+    code = cli_module.main(["Open amazon.de"])
+    assert code == 0
+
+    out = capsys.readouterr().out
+    payload = json.loads(out)
+    assert payload["response"]["return"] == "Task completed successfully"
+    assert payload["response"]["data"] == "file1.txt\nfile2.txt"
+    assert payload["return"] == "Task completed successfully"
+    assert payload["data"] == "file1.txt\nfile2.txt"