feat: add challenge and red-blue competitions across API and web

api/tests/test_containers_integration_tests/test_web_challenges.py

@@ -0,0 +1,14 @@
+from __future__ import annotations
+
+from flask.testing import FlaskClient
+
+
+class TestWebChallenges:
+    def test_list_and_detail(self, test_client_with_containers: FlaskClient):
+        # list
+        resp = test_client_with_containers.get("/api/web/challenges")
+        assert resp.status_code == 200
+        data = resp.get_json()
+        assert data["result"] == "success"
+
+

api/tests/test_containers_integration_tests/test_web_red_blue_challenges.py

@@ -0,0 +1,13 @@
+from __future__ import annotations
+
+from flask.testing import FlaskClient
+
+
+class TestWebRedBlueChallenges:
+    def test_list(self, test_client_with_containers: FlaskClient):
+        resp = test_client_with_containers.get("/api/web/red-blue-challenges")
+        assert resp.status_code == 200
+        data = resp.get_json()
+        assert data["result"] == "success"
+
+

api/tests/unit_tests/services/test_challenge_scorer_service.py

@@ -0,0 +1,144 @@
+"""Tests for ChallengeScorerService."""
+
+from __future__ import annotations
+
+import pytest
+
+from services.challenge_scorer_service import ChallengeScorerService
+
+
+class TestChallengeScorerService:
+    """Test custom scorer plugin loading and execution."""
+
+    def test_weighted_scorer_success(self):
+        """Test WeightedScorer with successful attempt."""
+        metrics = {
+            "succeeded": True,
+            "tokens_total": 1000,
+            "elapsed_ms": 5000,  # 5 seconds
+            "rating": 8,
+            "created_at": 1730000000000,
+        }
+
+        config = {
+            "success_bonus": 100.0,
+            "rating_weight": 10.0,
+            "time_penalty": 1.0,
+            "token_penalty": 0.01,
+        }
+
+        ctx = {
+            "tenant_id": "test-tenant",
+            "app_id": "test-app",
+            "workflow_id": "test-workflow",
+            "challenge_id": "test-challenge",
+            "timeout_ms": 5000,
+        }
+
+        result = ChallengeScorerService.score_with_plugin(
+            scorer_plugin_id="builtin.weighted_scorer",
+            scorer_entrypoint="services.scorers.weighted:WeightedScorer",
+            metrics=metrics,
+            config=config,
+            ctx=ctx,
+        )
+
+        # Expected: 100 (success) + 80 (8*10 rating) - 5 (5s*1.0) - 10 (1000*0.01) = 165
+        assert result["score"] == 165.0
+        assert "details" in result
+        assert result["details"]["base"] == 100.0
+        assert result["details"]["rating_contribution"] == 80.0
+        assert result["details"]["time_penalty"] == 5.0
+        assert result["details"]["token_penalty"] == 10.0
+
+    def test_weighted_scorer_failure(self):
+        """Test WeightedScorer with failed attempt."""
+        metrics = {
+            "succeeded": False,
+            "tokens_total": 500,
+            "elapsed_ms": 2000,  # 2 seconds
+            "rating": 3,
+            "created_at": 1730000000000,
+        }
+
+        config = {
+            "success_bonus": 100.0,
+            "rating_weight": 10.0,
+            "time_penalty": 1.0,
+            "token_penalty": 0.01,
+        }
+
+        ctx = {
+            "tenant_id": "test-tenant",
+            "app_id": "test-app",
+            "challenge_id": "test-challenge",
+            "timeout_ms": 5000,
+        }
+
+        result = ChallengeScorerService.score_with_plugin(
+            scorer_plugin_id="builtin.weighted_scorer",
+            scorer_entrypoint="services.scorers.weighted:WeightedScorer",
+            metrics=metrics,
+            config=config,
+            ctx=ctx,
+        )
+
+        # Expected: 0 (no success bonus) + 30 (3*10) - 2 (2s*1.0) - 5 (500*0.01) = 23
+        assert result["score"] == 23.0
+
+    def test_weighted_scorer_minimum_zero(self):
+        """Test WeightedScorer never returns negative scores."""
+        metrics = {
+            "succeeded": False,
+            "tokens_total": 10000,  # High token count
+            "elapsed_ms": 30000,  # 30 seconds
+            "rating": 1,
+            "created_at": 1730000000000,
+        }
+
+        config = {
+            "success_bonus": 100.0,
+            "rating_weight": 10.0,
+            "time_penalty": 1.0,
+            "token_penalty": 0.01,
+        }
+
+        ctx = {
+            "tenant_id": "test-tenant",
+            "app_id": "test-app",
+            "challenge_id": "test-challenge",
+            "timeout_ms": 5000,
+        }
+
+        result = ChallengeScorerService.score_with_plugin(
+            scorer_plugin_id="builtin.weighted_scorer",
+            scorer_entrypoint="services.scorers.weighted:WeightedScorer",
+            metrics=metrics,
+            config=config,
+            ctx=ctx,
+        )
+
+        # Expected: 0 + 10 - 30 - 100 = -120, but clamped to 0
+        assert result["score"] == 0.0
+
+    def test_scorer_with_missing_plugin(self):
+        """Test error handling for missing plugin."""
+        with pytest.raises(ValueError, match="Failed to load scorer plugin"):
+            ChallengeScorerService.score_with_plugin(
+                scorer_plugin_id="nonexistent",
+                scorer_entrypoint="nonexistent.module:NonexistentScorer",
+                metrics={},
+                config={},
+                ctx={"timeout_ms": 5000},
+            )
+
+    def test_scorer_with_invalid_entrypoint(self):
+        """Test error handling for invalid entrypoint format."""
+        with pytest.raises(ValueError, match="scorer_plugin_id and scorer_entrypoint are required"):
+            ChallengeScorerService.score_with_plugin(
+                scorer_plugin_id=None,
+                scorer_entrypoint=None,
+                metrics={},
+                config={},
+                ctx={"timeout_ms": 5000},
+            )

api/tests/unit_tests/services/test_challenge_service.py

@@ -0,0 +1,40 @@
+from __future__ import annotations
+
+from models.challenge import ChallengeAttempt
+from services.challenge_service import ChallengeService
+
+
+def test_evaluate_outcome_regex_match():
+    ok, details = ChallengeService.evaluate_outcome(
+        "Hello SECRET",
+        {"success_type": "regex", "success_pattern": "secret"},
+    )
+    assert ok is True
+    assert details.get("mode") == "regex"
+
+
+def test_evaluate_outcome_contains():
+    ok, _ = ChallengeService.evaluate_outcome(
+        "hello world",
+        {"success_type": "contains", "success_pattern": "world"},
+    )
+    assert ok is True
+
+
+def test_record_attempt_creates_row(mocker):
+    # mock db.session
+    session = mocker.MagicMock()
+    attempt = ChallengeService.record_attempt(
+        tenant_id="t1",
+        challenge_id="c1",
+        end_user_id=None,
+        account_id=None,
+        workflow_run_id=None,
+        succeeded=True,
+        score=10.0,
+        session=session,
+    )
+    assert isinstance(attempt, ChallengeAttempt)
+    session.add.assert_called_once()
+    session.commit.assert_called_once()
+

api/tests/unit_tests/services/test_red_blue_service.py

@@ -0,0 +1,34 @@
+from __future__ import annotations
+
+from types import SimpleNamespace
+
+from services.red_blue_service import RedBlueService
+
+
+def test_submit_prompt_creates_submission(mocker):
+    session = mocker.MagicMock()
+    sub = RedBlueService.submit_prompt(
+        challenge_id="cid",
+        tenant_id="tid",
+        team="red",
+        prompt="attack",
+        account_id="aid",
+        end_user_id="eid",
+        session=session,
+    )
+    assert sub.team == "red"
+    session.add.assert_called_once()
+    session.commit.assert_called_once()
+
+
+def test_select_counterparty_submission_latest_active(mocker):
+    c = SimpleNamespace(id="cid")
+    session = mocker.MagicMock()
+    qs = (
+        session.query.return_value.filter.return_value.order_by.return_value
+    )
+    qs.first.return_value = SimpleNamespace(id="subid", team="blue")
+    sub = RedBlueService.select_counterparty_submission(challenge=c, team="red", session=session)
+    assert sub.team == "blue"
+
+