feat(conductor): Add retry loop for Tech Lead JSON parsing

conductor/tracks/robust_json_parsing_tech_lead_20260302/plan.md

@@ -3,13 +3,13 @@
 > **TEST DEBT FIX:** Due to ongoing test architecture instability (documented in `test_architecture_integrity_audit_20260304`), do NOT write new `live_gui` integration tests for this track. Rely strictly on in-process `unittest.mock` for the `ai_client` to verify the retry logic.
 
 ## Phase 1: Implementation of Retry Logic
-- [ ] Task: Initialize MMA Environment `activate_skill mma-orchestrator`
-- [ ] Task: Implement Retry Loop in `generate_tickets`
+- [x] Task: Initialize MMA Environment `activate_skill mma-orchestrator`
+- [x] Task: Implement Retry Loop in `generate_tickets`
     - [ ] WHERE: `conductor_tech_lead.py:generate_tickets`
     - [ ] WHAT: Wrap the `send` and `json.loads` calls in a `for _ in range(max_retries)` loop.
     - [ ] HOW: If `JSONDecodeError` is caught, append an error message to the context and loop. If it succeeds, `break` and return.
     - [ ] SAFETY: Ensure token limits aren't massively breached by appending huge error states. Truncate raw output if necessary.
-- [ ] Task: Conductor - User Manual Verification 'Phase 1: Implementation' (Protocol in workflow.md)
+- [x] Task: Conductor - User Manual Verification 'Phase 1: Implementation' (Protocol in workflow.md)
 
 ## Phase 2: Unit Testing
 - [ ] Task: Write Simulation Tests for JSON Parsing

src/conductor_tech_lead.py

@@ -21,6 +21,9 @@ def generate_tickets(track_brief: str, module_skeletons: str) -> list[dict[str,
  old_system_prompt = ai_client._custom_system_prompt
  ai_client.set_custom_system_prompt(system_prompt or "")
  ai_client.current_tier = "Tier 2"
+ last_error = None
+ try:
+  for _ in range(3):
    try:
    # 3. Call Tier 2 Model
     response = ai_client.send(
@@ -41,9 +44,12 @@ def generate_tickets(track_brief: str, module_skeletons: str) -> list[dict[str,
       json_match = match.group(0)
     tickets: list[dict[str, Any]] = json.loads(json_match)
     return tickets
- except Exception as e:
-  print(f"Error parsing Tier 2 response: {e}")
-  return []
+   except json.JSONDecodeError as e:
+    last_error = e
+    correction = f"\n\nYour previous output failed to parse as JSON: {e}. Here was your raw output:\n{json_match[:500]}\n\nPlease fix the formatting and output ONLY valid JSON array."
+    user_message += correction
+    print(f"JSON parsing error, retrying... ({_ + 1}/3)")
+  raise RuntimeError(f"Failed to generate valid JSON tickets after 3 attempts. Last error: {last_error}")
  finally:
  # Restore old system prompt and clear tier tag
   ai_client.set_custom_system_prompt(old_system_prompt or "")

tests/test_conductor_tech_lead.py

@@ -7,9 +7,9 @@ class TestConductorTechLead(unittest.TestCase):
  def test_generate_tickets_parse_error(self) -> None:
   with patch('src.ai_client.send') as mock_send:
    mock_send.return_value = "invalid json"
-            # conductor_tech_lead.generate_tickets returns [] on error, doesn't raise
-            tickets = conductor_tech_lead.generate_tickets("brief", "skeletons")
-            self.assertEqual(tickets, [])
+   # conductor_tech_lead.generate_tickets now raises RuntimeError on error
+   with pytest.raises(RuntimeError):
+    conductor_tech_lead.generate_tickets("brief", "skeletons")
 
  def test_generate_tickets_success(self) -> None:
   with patch('src.ai_client.send') as mock_send: