fix: restructure openenv.yaml tasks: from dict to flat list so platform grader check finds 30/30 tasks

ROOT CAUSE: competition platform does 'for t in yaml[tasks]' expecting a list.
Our tasks: was a dict with count/difficulties/definitions keys, so iteration
returned key strings (not task objects) -> 0 tasks found with grader -> FAIL.

FIX: tasks: is now a flat list of 30 task objects (each with task_id, difficulty,
grader). Metadata moved to task_count/task_difficulties/task_difficulty_weights
top-level keys. Platform now finds 30/30 tasks with grader.

Verified: 90/90 validate.py checks pass, openenv validate passes.

_test_grader_fix.py

@@ -0,0 +1,65 @@
+"""Test grader property and grade_episode changes."""
+import sys
+sys.path.insert(0, '.')
+from payops_env.tasks import TASKS
+from payops_env.grader import grade_episode
+
+# Test 1: task.grader property
+print('=== Test 1: task.grader property ===')
+missing = [t.task_id for t in TASKS if not hasattr(t, 'grader')]
+if missing:
+    print(f'FAIL: tasks missing grader property: {missing}')
+    sys.exit(1)
+else:
+    print(f'PASS: all {len(TASKS)} tasks have grader property')
+
+# Spot-check grader content
+t0 = TASKS[0]
+g = t0.grader
+assert 'type' in g and g['type'] == 'action_match', f'grader bad: {g}'
+assert 'correct_action' in g
+assert 'partial_credit' in g
+assert 'requires_investigation' in g
+assert 'regulatory_action' in g
+assert 'key_flags' in g
+print(f'PASS: grader property has required keys: {sorted(g.keys())}')
+
+# Test 2: grade_episode per_task_rewards have grader key
+print()
+print('=== Test 2: grade_episode per_task_rewards have grader key ===')
+sample_tasks = list(TASKS[:5])
+sample_actions = [t.correct_action for t in sample_tasks]
+result = grade_episode(sample_actions, sample_tasks)
+missing_gr = [pt['task_id'] for pt in result.per_task_rewards if 'grader' not in pt]
+if missing_gr:
+    print(f'FAIL: per_task_rewards entries missing grader key: {missing_gr}')
+    sys.exit(1)
+else:
+    print(f'PASS: all {len(result.per_task_rewards)} per_task_rewards entries have grader key')
+print(f'PASS: score={result.normalised_score}')
+
+# Test all 30 tasks
+result_all = grade_episode([t.correct_action for t in TASKS], list(TASKS))
+missing_all = [pt['task_id'] for pt in result_all.per_task_rewards if 'grader' not in pt]
+assert not missing_all, f'FAIL: {missing_all}'
+print(f'PASS: all 30 tasks graded with grader key in per_task_rewards')
+print(f'PASS: score={result_all.normalised_score}')
+
+# Test 3: openenv.yaml has task definitions with grader
+print()
+print('=== Test 3: openenv.yaml task definitions ===')
+import yaml
+with open('openenv.yaml') as f:
+    d = yaml.safe_load(f)
+# tasks: is now a flat list (changed from dict+definitions to list for platform compat)
+tasks_section = d.get('tasks', [])
+if isinstance(tasks_section, list):
+    defs = tasks_section
+else:
+    defs = tasks_section.get('definitions', [])
+tasks_with_grader = [t for t in defs if 'grader' in t]
+print(f'PASS: openenv.yaml has {len(defs)} task definitions, {len(tasks_with_grader)} with grader')
+assert len(tasks_with_grader) >= 3, f'FAIL: only {len(tasks_with_grader)} tasks with grader'
+
+print()
+print('ALL TESTS PASSED')

openenv.yaml

@@ -112,433 +112,435 @@ observation_space:
     - recent_decisions
     - network_graph
 
-# ── Tasks ─────────────────────────────────────────────────────────────────────
+# ── Tasks ───────────────────────────────────────────────────────────────────
+# Metadata kept separate so 'count: 30' still appears in file
+task_count: 30
+task_difficulties:
+  - easy
+  - medium
+  - hard
+  - critical
+task_difficulty_weights:
+  easy: 1.0
+  medium: 1.2
+  hard: 1.5
+  critical: 2.0
+
+# tasks: flat list — platform iterates this directly for grader check
 tasks:
-  count: 30
-  difficulties:
-    - easy
-    - medium
-    - hard
-    - critical
-  difficulty_weights:
-    easy: 1.0
-    medium: 1.2
-    hard: 1.5
-    critical: 2.0
-  definitions:
-    - task_id: EASY-001
-      difficulty: easy
-      grader:
-        type: action_match
-        correct_action: approve
-        partial_credit: {}
-        requires_investigation: []
-        regulatory_action: false
-        key_flags: []
-    - task_id: EASY-002
-      difficulty: easy
-      grader:
-        type: action_match
-        correct_action: reject
-        partial_credit:
-          escalate: 0.4
-        requires_investigation: []
-        regulatory_action: false
-        key_flags:
-          - sanctioned_country
-          - unknown_sender
-    - task_id: EASY-003
-      difficulty: easy
-      grader:
-        type: action_match
-        correct_action: approve
-        partial_credit: {}
-        requires_investigation: []
-        regulatory_action: false
-        key_flags: []
-    - task_id: EASY-004
-      difficulty: easy
-      grader:
-        type: action_match
-        correct_action: flag
-        partial_credit:
-          escalate: 0.6
-          hold: 0.5
-        requires_investigation: []
-        regulatory_action: false
-        key_flags:
-          - velocity_breach
-    - task_id: EASY-005
-      difficulty: easy
-      grader:
-        type: action_match
-        correct_action: approve
-        partial_credit: {}
-        requires_investigation: []
-        regulatory_action: false
-        key_flags: []
-    - task_id: EASY-006
-      difficulty: easy
-      grader:
-        type: action_match
-        correct_action: flag
-        partial_credit:
-          hold: 0.6
-        requires_investigation: []
-        regulatory_action: false
-        key_flags:
-          - duplicate_transaction
-    - task_id: MED-001
-      difficulty: medium
-      grader:
-        type: action_match
-        correct_action: escalate
-        partial_credit:
-          flag: 0.5
-          hold: 0.4
-        requires_investigation:
-          - inspect
-        regulatory_action: false
-        key_flags:
-          - cross_border
-          - high_value
-    - task_id: MED-002
-      difficulty: medium
-      grader:
-        type: action_match
-        correct_action: hold
-        partial_credit:
-          escalate: 0.6
-          flag: 0.4
-        requires_investigation:
-          - verify_kyc
-        regulatory_action: false
-        key_flags:
-          - kyc_expiry_90d
-    - task_id: MED-003
-      difficulty: medium
-      grader:
-        type: action_match
-        correct_action: flag
-        partial_credit:
-          hold: 0.5
-          escalate: 0.3
-        requires_investigation: []
-        regulatory_action: false
-        key_flags:
-          - amount_spike
-          - pattern_deviation
-    - task_id: MED-004
-      difficulty: medium
-      grader:
-        type: action_match
-        correct_action: flag
-        partial_credit:
-          escalate: 0.5
-          hold: 0.4
-        requires_investigation: []
-        regulatory_action: false
-        key_flags:
-          - crypto_exchange
-    - task_id: MED-005
-      difficulty: medium
-      grader:
-        type: action_match
-        correct_action: hold
-        partial_credit:
-          flag: 0.5
-          escalate: 0.4
-        requires_investigation:
-          - verify_kyc
-        regulatory_action: false
-        key_flags:
-          - kyc_expired
-    - task_id: MED-006
-      difficulty: medium
-      grader:
-        type: action_match
-        correct_action: escalate
-        partial_credit:
-          flag: 0.4
-          hold: 0.5
-        requires_investigation:
-          - request_docs
-        regulatory_action: false
-        key_flags:
-          - first_time_payee
-          - large_first_transfer
-    - task_id: MED-007
-      difficulty: medium
-      grader:
-        type: action_match
-        correct_action: hold
-        partial_credit:
-          escalate: 0.6
-          flag: 0.4
-        requires_investigation:
-          - contact_sender
-          - request_docs
-        regulatory_action: false
-        key_flags:
-          - bank_detail_change
-          - first_time_payee
-    - task_id: MED-008
-      difficulty: medium
-      grader:
-        type: action_match
-        correct_action: flag
-        partial_credit:
-          hold: 0.6
-          escalate: 0.4
-        requires_investigation:
-          - verify_kyc
-        regulatory_action: false
-        key_flags:
-          - thin_credit_file
-          - new_account
-    - task_id: HARD-001
-      difficulty: hard
-      grader:
-        type: action_match
-        correct_action: escalate
-        partial_credit:
-          flag: 0.6
-          hold: 0.5
-          reject: 0.3
-        requires_investigation:
-          - inspect
-        regulatory_action: false
-        key_flags:
-          - solicitor_mule_pattern
-    - task_id: HARD-002
-      difficulty: hard
-      grader:
-        type: action_match
-        correct_action: reject
-        partial_credit:
-          escalate: 0.6
-          hold: 0.5
-          flag: 0.3
-        requires_investigation:
-          - contact_sender
-        regulatory_action: false
-        key_flags:
-          - app_scam_indicator
-          - mule_account_pattern
-    - task_id: HARD-003
-      difficulty: hard
-      grader:
-        type: action_match
-        correct_action: reject
-        partial_credit:
-          escalate: 0.5
-          flag: 0.2
-        requires_investigation:
-          - inspect
-        regulatory_action: true
-        key_flags:
-          - structuring_pattern
-          - ctr_threshold_avoidance
-    - task_id: HARD-004
-      difficulty: hard
-      grader:
-        type: action_match
-        correct_action: approve
-        partial_credit:
-          escalate: 0.5
-          hold: 0.4
-          flag: 0.3
-        requires_investigation:
-          - inspect
-        regulatory_action: false
-        key_flags:
-          - fx_settlement
-    - task_id: HARD-005
-      difficulty: hard
-      grader:
-        type: action_match
-        correct_action: escalate
-        partial_credit:
-          flag: 0.5
-          hold: 0.4
-          reject: 0.3
-        requires_investigation:
-          - contact_sender
-          - inspect
-        regulatory_action: false
-        key_flags:
-          - internal_to_personal
-          - unusual_beneficiary
-    - task_id: HARD-006
-      difficulty: hard
-      grader:
-        type: action_match
-        correct_action: flag
-        partial_credit:
-          hold: 0.6
-          escalate: 0.5
-        requires_investigation:
-          - inspect
-        regulatory_action: false
-        key_flags:
-          - dormant_receiver
-          - sudden_activity
-    - task_id: HARD-007
-      difficulty: hard
-      grader:
-        type: action_match
-        correct_action: reject
-        partial_credit:
-          hold: 0.5
-          escalate: 0.4
-          flag: 0.3
-        requires_investigation:
-          - contact_sender
-          - inspect
-        regulatory_action: false
-        key_flags:
-          - sim_swap_detected
-          - new_withdrawal_address
-    - task_id: HARD-008
-      difficulty: hard
-      grader:
-        type: action_match
-        correct_action: reject
-        partial_credit:
-          hold: 0.5
-          escalate: 0.4
-          flag: 0.3
-        requires_investigation:
-          - contact_sender
-          - inspect
-        regulatory_action: false
-        key_flags:
-          - romance_scam_indicator
-          - escalating_transfers
-    - task_id: HARD-009
-      difficulty: hard
-      grader:
-        type: action_match
-        correct_action: escalate
-        partial_credit:
-          hold: 0.5
-          flag: 0.4
-          reject: 0.3
-        requires_investigation:
-          - inspect
-          - verify_kyc
-        regulatory_action: false
-        key_flags:
-          - new_account_10d
-          - perfect_score_anomaly
-    - task_id: HARD-010
-      difficulty: hard
-      grader:
-        type: action_match
-        correct_action: reject
-        partial_credit:
-          hold: 0.5
-          escalate: 0.4
-          flag: 0.3
-        requires_investigation:
-          - contact_sender
-          - inspect
-        regulatory_action: false
-        key_flags:
-          - direct_deposit_changed
-          - payroll_diversion_indicator
-    - task_id: CRIT-001
-      difficulty: critical
-      grader:
-        type: action_match
-        correct_action: approve
-        partial_credit:
-          escalate: 0.5
-          hold: 0.4
-          flag: 0.3
-        requires_investigation:
-          - inspect
-          - request_docs
-        regulatory_action: false
-        key_flags:
-          - first_time_payee
-          - new_counterparty
-    - task_id: CRIT-002
-      difficulty: critical
-      grader:
-        type: action_match
-        correct_action: reject
-        partial_credit:
-          escalate: 0.4
-          flag: 0.2
-        requires_investigation:
-          - inspect
-        regulatory_action: true
-        key_flags:
-          - fraud_ring_indicator
-          - coordinated_transfers
-    - task_id: CRIT-003
-      difficulty: critical
-      grader:
-        type: action_match
-        correct_action: escalate
-        partial_credit:
-          flag: 0.4
-          hold: 0.35
-          reject: 0.3
-        requires_investigation:
-          - inspect
-          - request_docs
-        regulatory_action: true
-        key_flags:
-          - invoice_mismatch
-          - trade_finance
-    - task_id: CRIT-004
-      difficulty: critical
-      grader:
-        type: action_match
-        correct_action: reject
-        partial_credit:
-          escalate: 0.5
-          hold: 0.4
-        requires_investigation:
-          - contact_sender
-          - inspect
-        regulatory_action: false
-        key_flags:
-          - geo_impossible_login
-          - account_takeover_indicator
-    - task_id: CRIT-005
-      difficulty: critical
-      grader:
-        type: action_match
-        correct_action: reject
-        partial_credit:
-          escalate: 0.5
-          hold: 0.4
-        requires_investigation:
-          - inspect
-          - request_docs
-          - verify_kyc
-        regulatory_action: true
-        key_flags:
-          - shell_company_indicator
-          - sanctions_adjacent
-    - task_id: CRIT-006
-      difficulty: critical
-      grader:
-        type: action_match
-        correct_action: escalate
-        partial_credit:
-          hold: 0.5
-          reject: 0.4
-        requires_investigation:
-          - inspect
-          - request_docs
-        regulatory_action: true
-        key_flags:
-          - fincen_311_alert
-          - correspondent_risk
+  - task_id: EASY-001
+    difficulty: easy
+    grader:
+      type: action_match
+      correct_action: approve
+      partial_credit: {}
+      requires_investigation: []
+      regulatory_action: false
+      key_flags: []
+  - task_id: EASY-002
+    difficulty: easy
+    grader:
+      type: action_match
+      correct_action: reject
+      partial_credit:
+        escalate: 0.4
+      requires_investigation: []
+      regulatory_action: false
+      key_flags:
+        - sanctioned_country
+        - unknown_sender
+  - task_id: EASY-003
+    difficulty: easy
+    grader:
+      type: action_match
+      correct_action: approve
+      partial_credit: {}
+      requires_investigation: []
+      regulatory_action: false
+      key_flags: []
+  - task_id: EASY-004
+    difficulty: easy
+    grader:
+      type: action_match
+      correct_action: flag
+      partial_credit:
+        escalate: 0.6
+        hold: 0.5
+      requires_investigation: []
+      regulatory_action: false
+      key_flags:
+        - velocity_breach
+  - task_id: EASY-005
+    difficulty: easy
+    grader:
+      type: action_match
+      correct_action: approve
+      partial_credit: {}
+      requires_investigation: []
+      regulatory_action: false
+      key_flags: []
+  - task_id: EASY-006
+    difficulty: easy
+    grader:
+      type: action_match
+      correct_action: flag
+      partial_credit:
+        hold: 0.6
+      requires_investigation: []
+      regulatory_action: false
+      key_flags:
+        - duplicate_transaction
+  - task_id: MED-001
+    difficulty: medium
+    grader:
+      type: action_match
+      correct_action: escalate
+      partial_credit:
+        flag: 0.5
+        hold: 0.4
+      requires_investigation:
+        - inspect
+      regulatory_action: false
+      key_flags:
+        - cross_border
+        - high_value
+  - task_id: MED-002
+    difficulty: medium
+    grader:
+      type: action_match
+      correct_action: hold
+      partial_credit:
+        escalate: 0.6
+        flag: 0.4
+      requires_investigation:
+        - verify_kyc
+      regulatory_action: false
+      key_flags:
+        - kyc_expiry_90d
+  - task_id: MED-003
+    difficulty: medium
+    grader:
+      type: action_match
+      correct_action: flag
+      partial_credit:
+        hold: 0.5
+        escalate: 0.3
+      requires_investigation: []
+      regulatory_action: false
+      key_flags:
+        - amount_spike
+        - pattern_deviation
+  - task_id: MED-004
+    difficulty: medium
+    grader:
+      type: action_match
+      correct_action: flag
+      partial_credit:
+        escalate: 0.5
+        hold: 0.4
+      requires_investigation: []
+      regulatory_action: false
+      key_flags:
+        - crypto_exchange
+  - task_id: MED-005
+    difficulty: medium
+    grader:
+      type: action_match
+      correct_action: hold
+      partial_credit:
+        flag: 0.5
+        escalate: 0.4
+      requires_investigation:
+        - verify_kyc
+      regulatory_action: false
+      key_flags:
+        - kyc_expired
+  - task_id: MED-006
+    difficulty: medium
+    grader:
+      type: action_match
+      correct_action: escalate
+      partial_credit:
+        flag: 0.4
+        hold: 0.5
+      requires_investigation:
+        - request_docs
+      regulatory_action: false
+      key_flags:
+        - first_time_payee
+        - large_first_transfer
+  - task_id: MED-007
+    difficulty: medium
+    grader:
+      type: action_match
+      correct_action: hold
+      partial_credit:
+        escalate: 0.6
+        flag: 0.4
+      requires_investigation:
+        - contact_sender
+        - request_docs
+      regulatory_action: false
+      key_flags:
+        - bank_detail_change
+        - first_time_payee
+  - task_id: MED-008
+    difficulty: medium
+    grader:
+      type: action_match
+      correct_action: flag
+      partial_credit:
+        hold: 0.6
+        escalate: 0.4
+      requires_investigation:
+        - verify_kyc
+      regulatory_action: false
+      key_flags:
+        - thin_credit_file
+        - new_account
+  - task_id: HARD-001
+    difficulty: hard
+    grader:
+      type: action_match
+      correct_action: escalate
+      partial_credit:
+        flag: 0.6
+        hold: 0.5
+        reject: 0.3
+      requires_investigation:
+        - inspect
+      regulatory_action: false
+      key_flags:
+        - solicitor_mule_pattern
+  - task_id: HARD-002
+    difficulty: hard
+    grader:
+      type: action_match
+      correct_action: reject
+      partial_credit:
+        escalate: 0.6
+        hold: 0.5
+        flag: 0.3
+      requires_investigation:
+        - contact_sender
+      regulatory_action: false
+      key_flags:
+        - app_scam_indicator
+        - mule_account_pattern
+  - task_id: HARD-003
+    difficulty: hard
+    grader:
+      type: action_match
+      correct_action: reject
+      partial_credit:
+        escalate: 0.5
+        flag: 0.2
+      requires_investigation:
+        - inspect
+      regulatory_action: true
+      key_flags:
+        - structuring_pattern
+        - ctr_threshold_avoidance
+  - task_id: HARD-004
+    difficulty: hard
+    grader:
+      type: action_match
+      correct_action: approve
+      partial_credit:
+        escalate: 0.5
+        hold: 0.4
+        flag: 0.3
+      requires_investigation:
+        - inspect
+      regulatory_action: false
+      key_flags:
+        - fx_settlement
+  - task_id: HARD-005
+    difficulty: hard
+    grader:
+      type: action_match
+      correct_action: escalate
+      partial_credit:
+        flag: 0.5
+        hold: 0.4
+        reject: 0.3
+      requires_investigation:
+        - contact_sender
+        - inspect
+      regulatory_action: false
+      key_flags:
+        - internal_to_personal
+        - unusual_beneficiary
+  - task_id: HARD-006
+    difficulty: hard
+    grader:
+      type: action_match
+      correct_action: flag
+      partial_credit:
+        hold: 0.6
+        escalate: 0.5
+      requires_investigation:
+        - inspect
+      regulatory_action: false
+      key_flags:
+        - dormant_receiver
+        - sudden_activity
+  - task_id: HARD-007
+    difficulty: hard
+    grader:
+      type: action_match
+      correct_action: reject
+      partial_credit:
+        hold: 0.5
+        escalate: 0.4
+        flag: 0.3
+      requires_investigation:
+        - contact_sender
+        - inspect
+      regulatory_action: false
+      key_flags:
+        - sim_swap_detected
+        - new_withdrawal_address
+  - task_id: HARD-008
+    difficulty: hard
+    grader:
+      type: action_match
+      correct_action: reject
+      partial_credit:
+        hold: 0.5
+        escalate: 0.4
+        flag: 0.3
+      requires_investigation:
+        - contact_sender
+        - inspect
+      regulatory_action: false
+      key_flags:
+        - romance_scam_indicator
+        - escalating_transfers
+  - task_id: HARD-009
+    difficulty: hard
+    grader:
+      type: action_match
+      correct_action: escalate
+      partial_credit:
+        hold: 0.5
+        flag: 0.4
+        reject: 0.3
+      requires_investigation:
+        - inspect
+        - verify_kyc
+      regulatory_action: false
+      key_flags:
+        - new_account_10d
+        - perfect_score_anomaly
+  - task_id: HARD-010
+    difficulty: hard
+    grader:
+      type: action_match
+      correct_action: reject
+      partial_credit:
+        hold: 0.5
+        escalate: 0.4
+        flag: 0.3
+      requires_investigation:
+        - contact_sender
+        - inspect
+      regulatory_action: false
+      key_flags:
+        - direct_deposit_changed
+        - payroll_diversion_indicator
+  - task_id: CRIT-001
+    difficulty: critical
+    grader:
+      type: action_match
+      correct_action: approve
+      partial_credit:
+        escalate: 0.5
+        hold: 0.4
+        flag: 0.3
+      requires_investigation:
+        - inspect
+        - request_docs
+      regulatory_action: false
+      key_flags:
+        - first_time_payee
+        - new_counterparty
+  - task_id: CRIT-002
+    difficulty: critical
+    grader:
+      type: action_match
+      correct_action: reject
+      partial_credit:
+        escalate: 0.4
+        flag: 0.2
+      requires_investigation:
+        - inspect
+      regulatory_action: true
+      key_flags:
+        - fraud_ring_indicator
+        - coordinated_transfers
+  - task_id: CRIT-003
+    difficulty: critical
+    grader:
+      type: action_match
+      correct_action: escalate
+      partial_credit:
+        flag: 0.4
+        hold: 0.35
+        reject: 0.3
+      requires_investigation:
+        - inspect
+        - request_docs
+      regulatory_action: true
+      key_flags:
+        - invoice_mismatch
+        - trade_finance
+  - task_id: CRIT-004
+    difficulty: critical
+    grader:
+      type: action_match
+      correct_action: reject
+      partial_credit:
+        escalate: 0.5
+        hold: 0.4
+      requires_investigation:
+        - contact_sender
+        - inspect
+      regulatory_action: false
+      key_flags:
+        - geo_impossible_login
+        - account_takeover_indicator
+  - task_id: CRIT-005
+    difficulty: critical
+    grader:
+      type: action_match
+      correct_action: reject
+      partial_credit:
+        escalate: 0.5
+        hold: 0.4
+      requires_investigation:
+        - inspect
+        - request_docs
+        - verify_kyc
+      regulatory_action: true
+      key_flags:
+        - shell_company_indicator
+        - sanctions_adjacent
+  - task_id: CRIT-006
+    difficulty: critical
+    grader:
+      type: action_match
+      correct_action: escalate
+      partial_credit:
+        hold: 0.5
+        reject: 0.4
+      requires_investigation:
+        - inspect
+        - request_docs
+      regulatory_action: true
+      key_flags:
+        - fincen_311_alert
+        - correspondent_risk
 
 # ── Budget ───────────────────────────────────────────────────────────────────
 budget: