WIP: Regression hell

config.toml

@@ -1,6 +1,6 @@
 [ai]
-provider = "gemini"
+provider = "deepseek"
-model = "gemini-2.5-flash-lite"
+model = "deepseek-v3"
 temperature = 0.0
 max_tokens = 8192
 history_trunc_limit = 8000

manualslop_layout.ini

@@ -84,10 +84,10 @@ Size=900,700
 Collapsed=0
 
 [Window][Diagnostics]
-Pos=519,17
+Pos=2989,1760
-Size=1256,379
+Size=851,377
 Collapsed=0
-DockId=0x0000001A,0
+DockId=0x00000002,0
 
 [Window][Context Hub]
 Pos=0,17
@@ -102,16 +102,16 @@ Collapsed=0
 DockId=0x0000000D,0
 
 [Window][Discussion Hub]
-Pos=1777,17
+Pos=1750,17
-Size=1210,1883
+Size=1237,1270
 Collapsed=0
 DockId=0x00000013,0
 
 [Window][Operations Hub]
-Pos=519,398
+Pos=519,17
-Size=1256,1502
+Size=1229,1270
 Collapsed=0
-DockId=0x0000001B,0
+DockId=0x00000012,0
 
 [Window][Files & Media]
 Pos=0,979
@@ -132,15 +132,15 @@ Collapsed=0
 
 [Window][MMA Dashboard]
 Pos=2989,17
-Size=851,2120
+Size=851,1741
 Collapsed=0
-DockId=0x00000004,0
+DockId=0x00000001,0
 
 [Window][Log Management]
 Pos=2989,17
-Size=851,2120
+Size=851,1741
 Collapsed=0
-DockId=0x00000004,1
+DockId=0x00000001,1
 
 [Window][Track Proposal]
 Pos=709,326
@@ -148,26 +148,26 @@ Size=262,209
 Collapsed=0
 
 [Window][Tier 1: Strategy]
-Pos=519,1902
+Pos=519,1289
-Size=1836,235
+Size=513,848
 Collapsed=0
 DockId=0x00000014,0
 
 [Window][Tier 2: Tech Lead]
-Pos=2357,1902
+Pos=1034,1289
-Size=253,235
+Size=714,848
 Collapsed=0
 DockId=0x00000016,0
 
 [Window][Tier 4: QA]
-Pos=2803,1902
+Pos=2576,1289
-Size=184,235
+Size=411,848
 Collapsed=0
 DockId=0x00000019,0
 
 [Window][Tier 3: Workers]
-Pos=2612,1902
+Pos=1750,1289
-Size=189,235
+Size=824,848
 Collapsed=0
 DockId=0x00000018,0
 
@@ -265,20 +265,20 @@ DockSpace               ID=0xAFC85805 Window=0x079D3A04 Pos=0,17 Size=3840,2120
         DockNode        ID=0x00000005 Parent=0x00000007 SizeRef=295,960 Selected=0xF4139CA2
         DockNode        ID=0x00000006 Parent=0x00000007 SizeRef=295,1158 CentralNode=1 Selected=0x7BD57D6A
       DockNode          ID=0x0000000E Parent=0x0000000B SizeRef=2468,858 Split=Y Selected=0x418C7449
-        DockNode        ID=0x00000010 Parent=0x0000000E SizeRef=868,1883 Split=X Selected=0x418C7449
+        DockNode        ID=0x00000010 Parent=0x0000000E SizeRef=868,1270 Split=X Selected=0x418C7449
-          DockNode      ID=0x00000012 Parent=0x00000010 SizeRef=1256,402 Split=Y Selected=0xB4CBF21A
+          DockNode      ID=0x00000012 Parent=0x00000010 SizeRef=1229,402 Selected=0x418C7449
-            DockNode    ID=0x0000001A Parent=0x00000012 SizeRef=1141,379 Selected=0xB4CBF21A
+          DockNode      ID=0x00000013 Parent=0x00000010 SizeRef=1237,402 Selected=0x6F2B5B04
-            DockNode    ID=0x0000001B Parent=0x00000012 SizeRef=1141,1502 Selected=0x418C7449
+        DockNode        ID=0x00000011 Parent=0x0000000E SizeRef=868,848 Split=X Selected=0x5CDB7A4B
-          DockNode      ID=0x00000013 Parent=0x00000010 SizeRef=1210,402 Selected=0x6F2B5B04
+          DockNode      ID=0x00000014 Parent=0x00000011 SizeRef=513,837 Selected=0xBB346584
-        DockNode        ID=0x00000011 Parent=0x0000000E SizeRef=868,235 Split=X Selected=0x5CDB7A4B
+          DockNode      ID=0x00000015 Parent=0x00000011 SizeRef=1953,837 Split=X Selected=0x5CDB7A4B
-          DockNode      ID=0x00000014 Parent=0x00000011 SizeRef=1836,837 Selected=0xBB346584
+            DockNode    ID=0x00000016 Parent=0x00000015 SizeRef=714,837 Selected=0x390E7942
-          DockNode      ID=0x00000015 Parent=0x00000011 SizeRef=630,837 Split=X Selected=0x5CDB7A4B
+            DockNode    ID=0x00000017 Parent=0x00000015 SizeRef=1237,837 Split=X Selected=0x655BC6E9
-            DockNode    ID=0x00000016 Parent=0x00000015 SizeRef=730,837 Selected=0x390E7942
+              DockNode  ID=0x00000018 Parent=0x00000017 SizeRef=824,874 Selected=0x655BC6E9
-            DockNode    ID=0x00000017 Parent=0x00000015 SizeRef=1083,837 Split=X Selected=0x655BC6E9
+              DockNode  ID=0x00000019 Parent=0x00000017 SizeRef=411,874 Selected=0x5CDB7A4B
-              DockNode  ID=0x00000018 Parent=0x00000017 SizeRef=547,874 Selected=0x655BC6E9
-              DockNode  ID=0x00000019 Parent=0x00000017 SizeRef=534,874 Selected=0x5CDB7A4B
     DockNode            ID=0x0000000D Parent=0x00000003 SizeRef=435,1186 Selected=0x363E93D6
-  DockNode              ID=0x00000004 Parent=0xAFC85805 SizeRef=851,1183 Selected=0x3AEC3498
+  DockNode              ID=0x00000004 Parent=0xAFC85805 SizeRef=851,1183 Split=Y Selected=0x3AEC3498
+    DockNode            ID=0x00000001 Parent=0x00000004 SizeRef=851,1741 Selected=0x3AEC3498
+    DockNode            ID=0x00000002 Parent=0x00000004 SizeRef=851,377 Selected=0xB4CBF21A
 
 ;;;<<<Layout_655921752_Default>>>;;;
 ;;;<<<HelloImGui_Misc>>>;;;

mock_debug_prompt.txt

@@ -632,3 +632,71 @@ When you need to verify a change, rely on the exit code and stdout/stderr from t
 
 1+ 1?
 ------------------
+--- MOCK INVOKED ---
+ARGS: ['C:\\projects\\manual_slop\\tests\\mock_gemini_cli.py', '-m', 'gemini-2.5-flash-lite', '--prompt', '', '--output-format', 'stream-json']
+PROMPT:
+You are a helpful coding assistant with access to a PowerShell tool (run_powershell) and MCP tools (file access: read_file, list_directory, search_files, get_file_summary, web access: web_search, fetch_url). When calling file/directory tools, always use the 'path' parameter for the target path. When asked to create or edit files, prefer targeted edits over full rewrites. Always explain what you are doing before invoking the tool.
+
+When writing or rewriting large files (especially those containing quotes, backticks, or special characters), avoid python -c with inline strings. Instead: (1) write a .py helper script to disk using a PS here-string (@'...'@ for literal content), (2) run it with `python <script>`, (3) delete the helper. For small targeted edits, use PowerShell's (Get-Content) / .Replace() / Set-Content or Add-Content directly.
+
+When making function calls using tools that accept array or object parameters ensure those are structured using JSON. For example:
+When you need to verify a change, rely on the exit code and stdout/stderr from the tool — the user's context files are automatically refreshed after every tool call, so you do NOT need to re-read files that are already provided in the <context> block.
+
+<context>
+
+</context>
+
+[DISCUSSION HISTORY]
+
+## Discussion History
+
+### Discussion Excerpt 1
+
+@2026-03-06T20:40:49
+System:
+[PERFORMANCE ALERT] CPU usage high: 91.0%. Please consider optimizing recent changes or reducing load.
+
+---
+
+### Discussion Excerpt 2
+
+@2026-03-06T20:47:26
+System:
+[PERFORMANCE ALERT] CPU usage high: 95.5%. Please consider optimizing recent changes or reducing load.
+
+---
+
+### Discussion Excerpt 3
+
+@2026-03-06T20:48:08
+System:
+[PERFORMANCE ALERT] CPU usage high: 93.4%. Please consider optimizing recent changes or reducing load.
+
+---
+
+### Discussion Excerpt 4
+
+@2026-03-06T20:49:49
+System:
+[PERFORMANCE ALERT] CPU usage high: 85.4%. Please consider optimizing recent changes or reducing load.
+
+---
+
+### Discussion Excerpt 5
+
+@2026-03-06T20:51:47
+System:
+[PERFORMANCE ALERT] CPU usage high: 86.7%. Please consider optimizing recent changes or reducing load.
+
+---
+
+### Discussion Excerpt 6
+
+@2026-03-06T20:53:31
+System:
+[PERFORMANCE ALERT] CPU usage high: 96.6%. Please consider optimizing recent changes or reducing load.
+
+---
+
+testing gemini cli
+------------------

src/ai_client.py

@@ -252,7 +252,12 @@ def _classify_gemini_error(exc: Exception) -> ProviderError:
  return ProviderError("unknown", "gemini", exc)
 
 def _classify_deepseek_error(exc: Exception) -> ProviderError:
- body = str(exc).lower()
+ body = ""
+ if isinstance(exc, requests.exceptions.HTTPError) and exc.response is not None:
+  body = exc.response.text.lower()
+ else:
+  body = str(exc).lower()
+ 
  if "429" in body or "rate" in body:
   return ProviderError("rate_limit", "deepseek", exc)
  if "401" in body or "403" in body or "auth" in body or "api key" in body:
@@ -263,6 +268,13 @@ def _classify_deepseek_error(exc: Exception) -> ProviderError:
   return ProviderError("quota", "deepseek", exc)
  if "connection" in body or "timeout" in body or "network" in body:
   return ProviderError("network", "deepseek", exc)
+ 
+ if "400" in body or "bad request" in body:
+  # Try to wrap the original error with the response body for better debugging
+  if body:
+   new_exc = Exception(f"Bad Request (400): {body}")
+   return ProviderError("unknown", "deepseek", new_exc)
+
  return ProviderError("unknown", "deepseek", exc)
 
 def set_provider(provider: str, model: str) -> None:
@@ -637,6 +649,54 @@ def _build_file_diff_text(changed_items: list[dict[str, Any]]) -> str:
     parts.append(f"### `{path}` (no changes detected)")
  return "\n\n---\n\n".join(parts)
 
+def _build_deepseek_tools() -> list[dict[str, Any]]:
+ mcp_tools: list[dict[str, Any]] = []
+ for spec in mcp_client.MCP_TOOL_SPECS:
+  if _agent_tools.get(spec["name"], True):
+   mcp_tools.append({
+     "type": "function",
+     "function": {
+      "name":        spec["name"],
+      "description": spec["description"],
+      "parameters":  spec["parameters"],
+     }
+    })
+ tools_list = mcp_tools
+ if _agent_tools.get(TOOL_NAME, True):
+  powershell_tool: dict[str, Any] = {
+   "type": "function",
+   "function": {
+    "name": TOOL_NAME,
+    "description": (
+     "Run a PowerShell script within the project base_dir. "
+     "Use this to create, edit, rename, or delete files and directories. "
+     "The working directory is set to base_dir automatically. "
+     "Always prefer targeted edits over full rewrites where possible. "
+     "stdout and stderr are returned to you as the result."
+    ),
+    "parameters": {
+     "type": "object",
+     "properties": {
+      "script": {
+       "type": "string",
+       "description": "The PowerShell script to execute."
+      }
+     },
+     "required": ["script"]
+    }
+   }
+  }
+  tools_list.append(powershell_tool)
+ return tools_list
+
+_CACHED_DEEPSEEK_TOOLS: Optional[list[dict[str, Any]]] = None
+
+def _get_deepseek_tools() -> list[dict[str, Any]]:
+ global _CACHED_DEEPSEEK_TOOLS
+ if _CACHED_DEEPSEEK_TOOLS is None:
+  _CACHED_DEEPSEEK_TOOLS = _build_deepseek_tools()
+ return _CACHED_DEEPSEEK_TOOLS
+
 def _content_block_to_dict(block: Any) -> dict[str, Any]:
  if isinstance(block, dict):
   return block
@@ -1356,33 +1416,56 @@ def _send_deepseek(md_content: str, user_message: str, base_dir: str,
    "Authorization": f"Bearer {api_key}",
    "Content-Type": "application/json",
   }
-  current_api_messages: list[dict[str, Any]] = []
+  
+  is_reasoner = _model == "deepseek-reasoner"
+  
+  # Update history following Anthropic pattern
   with _deepseek_history_lock:
-   for msg in _deepseek_history:
+   if discussion_history and not _deepseek_history:
-    current_api_messages.append(msg)
+    user_content = f"[DISCUSSION HISTORY]\n\n{discussion_history}\n\n---\n\n{user_message}"
-  initial_user_message_content = user_message
+   else:
-  if discussion_history:
+    user_content = user_message
-   initial_user_message_content = f"[DISCUSSION HISTORY]\n\n{discussion_history}\n\n---\n\n{user_message}"
+   _deepseek_history.append({"role": "user", "content": user_content})
-  current_api_messages.append({"role": "user", "content": initial_user_message_content})
+  
-  request_payload: dict[str, Any] = {
-   "model": _model,
-   "messages": current_api_messages,
-   "temperature": _temperature,
-   "max_tokens": _max_tokens,
-   "stream": stream,
-  }
-  sys_msg = {"role": "system", "content": f"{_get_combined_system_prompt()}\n\n<context>\n{md_content}\n</context>"}
-  request_payload["messages"].insert(0, sys_msg)
   all_text_parts: list[str] = []
   _cumulative_tool_bytes = 0
-  round_idx = 0
+  
-  while round_idx <= MAX_TOOL_ROUNDS + 1:
+  for round_idx in range(MAX_TOOL_ROUNDS + 2):
+   current_api_messages: list[dict[str, Any]] = []
+   with _deepseek_history_lock:
+    for msg in _deepseek_history:
+     current_api_messages.append(msg)
+   
+   sys_msg = {"role": "system", "content": f"{_get_combined_system_prompt()}\n\n<context>\n{md_content}\n</context>"}
+   current_api_messages.insert(0, sys_msg)
+   
+   request_payload: dict[str, Any] = {
+    "model": _model,
+    "messages": current_api_messages,
+    "stream": stream,
+   }
+   
+   if not is_reasoner:
+    request_payload["temperature"] = _temperature
+    request_payload["max_tokens"] = _max_tokens
+    tools = _get_deepseek_tools()
+    if tools:
+     request_payload["tools"] = tools
+   
    events.emit("request_start", payload={"provider": "deepseek", "model": _model, "round": round_idx, "streaming": stream})
+   
    try:
-    response = requests.post(api_url, headers=headers, json=request_payload, timeout=60, stream=stream)
+    response = requests.post(api_url, headers=headers, json=request_payload, timeout=120, stream=stream)
     response.raise_for_status()
    except requests.exceptions.RequestException as e:
     raise _classify_deepseek_error(e) from e
+
+   assistant_text = ""
+   tool_calls_raw = []
+   reasoning_content = ""
+   finish_reason = "stop"
+   usage = {}
+
    if stream:
     aggregated_content = ""
     aggregated_tool_calls: list[dict[str, Any]] = []
@@ -1443,10 +1526,12 @@ def _send_deepseek(md_content: str, user_message: str, base_dir: str,
     reasoning_content = message.get("reasoning_content", "")
     finish_reason = choice.get("finish_reason", "stop")
     usage = response_data.get("usage", {})
+
    thinking_tags = ""
    if reasoning_content:
     thinking_tags = f"<thinking>\n{reasoning_content}\n</thinking>\n"
    full_assistant_text = thinking_tags + assistant_text
+   
    with _deepseek_history_lock:
     msg_to_store: dict[str, Any] = {"role": "assistant", "content": assistant_text}
     if reasoning_content:
@@ -1454,8 +1539,10 @@ def _send_deepseek(md_content: str, user_message: str, base_dir: str,
     if tool_calls_raw:
      msg_to_store["tool_calls"] = tool_calls_raw
     _deepseek_history.append(msg_to_store)
+   
    if full_assistant_text:
     all_text_parts.append(full_assistant_text)
+   
    _append_comms("IN", "response", {
      "round": round_idx,
      "stop_reason": finish_reason,
@@ -1464,6 +1551,7 @@ def _send_deepseek(md_content: str, user_message: str, base_dir: str,
      "usage": usage,
      "streaming": stream
     })
+   
    if finish_reason != "tool_calls" and not tool_calls_raw:
     break
    if round_idx > MAX_TOOL_ROUNDS:
@@ -1507,16 +1595,11 @@ def _send_deepseek(md_content: str, user_message: str, base_dir: str,
       "content": f"SYSTEM WARNING: Cumulative tool output exceeded {_MAX_TOOL_OUTPUT_BYTES // 1000}KB budget. Provide your final answer now."
      })
     _append_comms("OUT", "request", {"message": f"[TOOL OUTPUT BUDGET EXCEEDED: {_cumulative_tool_bytes} bytes]"})
+   
    with _deepseek_history_lock:
     for tr in tool_results_for_history:
      _deepseek_history.append(tr)
-   next_messages: list[dict[str, Any]] = []
+   
-   with _deepseek_history_lock:
-    for msg in _deepseek_history:
-     next_messages.append(msg)
-   next_messages.insert(0, sys_msg)
-   request_payload["messages"] = next_messages
-   round_idx += 1
   return "\n\n".join(all_text_parts) if all_text_parts else "(No text returned)"
  except Exception as e:
   raise _classify_deepseek_error(e) from e

src/api_hooks.py

@@ -2,6 +2,7 @@ from __future__ import annotations
 import json
 import threading
 import uuid
+import sys
 from http.server import ThreadingHTTPServer, BaseHTTPRequestHandler
 from typing import Any
 import logging
@@ -9,9 +10,15 @@ from src import session_logger
 
 def _get_app_attr(app: Any, name: str, default: Any = None) -> Any:
  if hasattr(app, name):
-  return getattr(app, name)
+  val = getattr(app, name)
+  sys.stderr.write(f"[DEBUG] _get_app_attr: found {name} in app -> {val}\n")
+  sys.stderr.flush()
+  return val
  if hasattr(app, 'controller') and hasattr(app.controller, name):
-  return getattr(app.controller, name)
+  val = getattr(app.controller, name)
+  sys.stderr.write(f"[DEBUG] _get_app_attr: found {name} in controller -> {val}\n")
+  sys.stderr.flush()
+  return val
  return default
 
 def _has_app_attr(app: Any, name: str) -> bool:
@@ -94,7 +101,10 @@ class HookHandler(BaseHTTPRequestHandler):
      settable = _get_app_attr(app, "_settable_fields", {})
      if field_tag in settable:
       attr = settable[field_tag]
-      result["value"] = _get_app_attr(app, attr, None)
+      val = _get_app_attr(app, attr, "MISSING")
+      sys.stderr.write(f"[DEBUG] Hook API: get_value {field_tag} -> attr {attr} -> {val}\n")
+      sys.stderr.flush()
+      result["value"] = val if val != "MISSING" else None
     finally: event.set()
    lock = _get_app_attr(app, "_pending_gui_tasks_lock")
    tasks = _get_app_attr(app, "_pending_gui_tasks")

src/app_controller.py

@@ -361,12 +361,16 @@ class AppController:
  def _process_pending_gui_tasks(self) -> None:
   if not self._pending_gui_tasks:
    return
+  sys.stderr.write(f"[DEBUG] _process_pending_gui_tasks: processing {len(self._pending_gui_tasks)} tasks\n")
+  sys.stderr.flush()
   with self._pending_gui_tasks_lock:
    tasks = self._pending_gui_tasks[:]
    self._pending_gui_tasks.clear()
   for task in tasks:
    try:
     action = task.get("action")
+    sys.stderr.write(f"[DEBUG] Processing GUI task: action={action}\n")
+    sys.stderr.flush()
     if action:
      session_logger.log_api_hook("PROCESS_TASK", action, str(task))
      # ...
@@ -428,14 +432,18 @@ class AppController:
      self._show_track_proposal_modal = True
     elif action == "mma_state_update":
     # Handle both internal (nested) and hook-server (flattened) payloads
-     payload = task.get("payload")
+     p = task.get("payload")
-     if not isinstance(payload, dict):
+     if not isinstance(p, dict):
-      payload = task # Fallback to task if payload missing or wrong type
+      p = task # Fallback to task itself if payload is missing or wrong type
-     self.mma_status = payload.get("status", "idle")
+     
-     self.active_tier = payload.get("active_tier")
+     sys.stderr.write(f"[DEBUG] mma_state_update: status={p.get('status')} active_tier={p.get('active_tier')}\n")
+     sys.stderr.flush()
+     
+     self.mma_status = p.get("status", self.mma_status)
+     self.active_tier = p.get("active_tier", self.active_tier)
      
      # Preserve existing model/provider config if not explicitly in payload
-     new_usage = payload.get("tier_usage", {})
+     new_usage = p.get("tier_usage", {})
      for tier, data in new_usage.items():
       if tier in self.mma_tier_usage:
        # Update usage counts but keep selected model/provider if not in update
@@ -446,12 +454,18 @@ class AppController:
       else:
        self.mma_tier_usage[tier] = data
 
-     self.active_tickets = payload.get("tickets", [])
+     self.active_tickets = p.get("tickets", [])
-     track_data = payload.get("track")
+     track_data = p.get("track")
      if track_data:
       tickets = []
       for t_data in self.active_tickets:
-       tickets.append(models.Ticket(**t_data))
+       if isinstance(t_data, models.Ticket):
+        tickets.append(t_data)
+       else:
+        # Map 'goal' from Godot format to 'description' if needed
+        if "goal" in t_data and "description" not in t_data:
+         t_data["description"] = t_data["goal"]
+        tickets.append(models.Ticket.from_dict(t_data))
       self.active_track = models.Track(
        id=track_data.get("id"),
        description=track_data.get("title", ""),

src/gui_2.py

@@ -1793,22 +1793,23 @@ class App:
   imgui.separator()
   # 4. Task DAG Visualizer
   imgui.text("Task DAG")
-  if self.active_track:
+  if self.active_track and self.node_editor_ctx:
    ed.set_current_editor(self.node_editor_ctx)
    ed.begin('Visual DAG')
    # Selection detection
    selected = ed.get_selected_nodes()
    if selected:
     for node_id in selected:
+     node_val = node_id.id()
      for t in self.active_tickets:
-      if abs(hash(str(t.get('id', '')))) == node_id:
+      if abs(hash(str(t.get('id', '')))) == node_val:
        self.ui_selected_ticket_id = str(t.get('id', ''))
        break
      break
    for t in self.active_tickets:
     tid = str(t.get('id', '??'))
     int_id = abs(hash(tid))
-    ed.begin_node(int_id)
+    ed.begin_node(ed.NodeId(int_id))
     imgui.text_colored(C_KEY, f"Ticket: {tid}")
     status = t.get('status', 'todo')
     s_col = C_VAL
@@ -1819,18 +1820,18 @@ class App:
     imgui.same_line()
     imgui.text_colored(s_col, status)
     imgui.text(f"Target: {t.get('target_file','')}")
-    ed.begin_pin(abs(hash(tid + "_in")), ed.PinKind.input)
+    ed.begin_pin(ed.PinId(abs(hash(tid + "_in"))), ed.PinKind.input)
     imgui.text("->")
     ed.end_pin()
     imgui.same_line()
-    ed.begin_pin(abs(hash(tid + "_out")), ed.PinKind.output)
+    ed.begin_pin(ed.PinId(abs(hash(tid + "_out"))), ed.PinKind.output)
     imgui.text("->")
     ed.end_pin()
     ed.end_node()
    for t in self.active_tickets:
     tid = str(t.get('id', '??'))
     for dep in t.get('depends_on', []):
-     ed.link(abs(hash(dep + "_" + tid)), abs(hash(dep + "_out")), abs(hash(tid + "_in")))
+     ed.link(ed.LinkId(abs(hash(dep + "_" + tid))), ed.PinId(abs(hash(dep + "_out"))), ed.PinId(abs(hash(tid + "_in"))))
    
    # Handle link creation
    if ed.begin_create():
@@ -1855,7 +1856,7 @@ class App:
           t.setdefault('depends_on', []).append(source_tid)
           self._push_mma_state_update()
          break
-   ed.end_create()
+     ed.end_create()
 
    # Handle link deletion
    if ed.begin_delete():
@@ -1870,7 +1871,7 @@ class App:
         t['depends_on'] = [dep for dep in deps if abs(hash(dep + "_" + tid)) != lid_val]
         self._push_mma_state_update()
         break
-   ed.end_delete()
+    ed.end_delete()
    # Validate DAG after any changes
    try:
     from src.dag_engine import TrackDAG
@@ -1881,7 +1882,6 @@ class App:
    except Exception:
     pass
    ed.end()
-   ed.set_current_editor(None)
    # 5. Add Ticket Form
    imgui.separator()
    if imgui.button("Add Ticket"):

src/models.py

@@ -1,5 +1,6 @@
 from __future__ import annotations
 import tomllib
+import datetime
 from dataclasses import dataclass, field
 from typing import List, Optional, Dict, Any, Union
 from pathlib import Path
@@ -179,12 +180,24 @@ class Metadata:
 
     @classmethod
     def from_dict(cls, data: Dict[str, Any]) -> "Metadata":
+        created = data.get("created_at")
+        updated = data.get("updated_at")
+        if isinstance(created, str):
+            try:
+                created = datetime.datetime.fromisoformat(created)
+            except ValueError:
+                pass
+        if isinstance(updated, str):
+            try:
+                updated = datetime.datetime.fromisoformat(updated)
+            except ValueError:
+                pass
         return cls(
             id=data["id"],
             name=data.get("name", ""),
             status=data.get("status", "todo"),
-            created_at=data.get("created_at"),
+            created_at=created,
-            updated_at=data.get("updated_at"),
+            updated_at=updated,
         )
 
 

tests/conftest.py

@@ -123,7 +123,7 @@ def mock_app() -> Generator[App, None, None]:
    'projects': {'paths': [], 'active': ''},
    'gui': {'show_windows': {}}
   }),
-  patch('src.gui_2.save_config'),
+  patch('src.models.save_config'),
   patch('src.gui_2.project_manager'),
   patch('src.gui_2.session_logger'),
   patch('src.gui_2.immapp.run'),
@@ -155,7 +155,7 @@ def app_instance() -> Generator[App, None, None]:
    'projects': {'paths': [], 'active': ''},
    'gui': {'show_windows': {}}
   }),
-  patch('src.gui_2.save_config'),
+  patch('src.models.save_config'),
   patch('src.gui_2.project_manager'),
   patch('src.gui_2.session_logger'),
   patch('src.gui_2.immapp.run'),

tests/test_conductor_engine_v2.py

@@ -34,10 +34,15 @@ def test_conductor_engine_run_executes_tickets_in_order(monkeypatch: pytest.Monk
  # Mocking lifecycle to mark ticket as complete so dependencies can be resolved
 
   def side_effect(ticket, context, *args, **kwargs):
-   ticket.mark_complete()
+   import threading
+   def do_work():
+    import time
+    time.sleep(0.1)
+    ticket.mark_complete()
+   threading.Thread(target=do_work).start()
    return "Success"
   mock_lifecycle.side_effect = side_effect
-  engine.run()
+  engine.run(max_ticks=20)
   
   vlogger.log_state("T1 Status Final", "todo", ticket1.status)
   vlogger.log_state("T2 Status Final", "todo", ticket2.status)
@@ -108,8 +113,8 @@ def test_run_worker_lifecycle_context_injection(monkeypatch: pytest.MonkeyPatch)
   run_worker_lifecycle(ticket, context, context_files=context_files)
   # Verify ASTParser calls: 
   # First file (primary) should get curated view, others (secondary) get skeleton
-  mock_ast_parser.get_curated_view.assert_called_once_with("def primary(): pass")
+  mock_ast_parser.get_curated_view.assert_called_once_with("def primary(): pass", path="primary.py")
-  mock_ast_parser.get_skeleton.assert_called_once_with("def secondary(): pass")
+  mock_ast_parser.get_skeleton.assert_called_once_with("def secondary(): pass", path="secondary.py")
   # Verify user_message contains the views
   _, kwargs = mock_send.call_args
   user_message = kwargs["user_message"]
@@ -238,10 +243,15 @@ def test_conductor_engine_dynamic_parsing_and_execution(monkeypatch: pytest.Monk
  # Mock run_worker_lifecycle to mark tickets as complete
  with patch("src.multi_agent_conductor.run_worker_lifecycle") as mock_lifecycle:
   def side_effect(ticket, context, *args, **kwargs):
-   ticket.mark_complete()
+   import threading
+   def do_work():
+    import time
+    time.sleep(0.1)
+    ticket.mark_complete()
+   threading.Thread(target=do_work).start()
    return "Success"
   mock_lifecycle.side_effect = side_effect
-  engine.run()
+  engine.run(max_ticks=20)
   assert mock_lifecycle.call_count == 3
   # Verify dependency order: T1 must be called before T2
   calls = [call[0][0].id for call in mock_lifecycle.call_args_list]

tests/test_deepseek_provider.py

@@ -111,3 +111,55 @@ def test_deepseek_streaming(mock_post: MagicMock) -> None:
   
   result = ai_client.send(md_content="Context", user_message="Stream test", base_dir=".", stream=True)
   assert result == "Hello World"
+
+@patch("requests.post")
+def test_deepseek_payload_verification(mock_post: MagicMock) -> None:
+ """
+    Verifies that the correct JSON payload (tools, history, params) is sent to DeepSeek.
+    """
+ ai_client.set_provider("deepseek", "deepseek-chat")
+ ai_client.reset_session()
+ with patch("src.ai_client._load_credentials", return_value={"deepseek": {"api_key": "test-key"}}):
+  mock_response = MagicMock()
+  mock_response.status_code = 200
+  mock_response.json.return_value = {
+   "choices": [{"message": {"content": "OK"}, "finish_reason": "stop"}]
+  }
+  mock_post.return_value = mock_response
+  
+  ai_client.send(md_content="Context", user_message="Message 1", base_dir=".", discussion_history="History")
+  
+  args, kwargs = mock_post.call_args
+  payload = kwargs["json"]
+  
+  assert payload["model"] == "deepseek-chat"
+  assert "tools" in payload
+  assert len(payload["messages"]) == 2 # system + user
+  assert "[DISCUSSION HISTORY]\n\nHistory" in payload["messages"][1]["content"]
+  assert "temperature" in payload
+  assert "max_tokens" in payload
+
+@patch("requests.post")
+def test_deepseek_reasoner_payload_verification(mock_post: MagicMock) -> None:
+ """
+    Verifies that deepseek-reasoner payload excludes tools and temperature.
+    """
+ ai_client.set_provider("deepseek", "deepseek-reasoner")
+ ai_client.reset_session()
+ with patch("src.ai_client._load_credentials", return_value={"deepseek": {"api_key": "test-key"}}):
+  mock_response = MagicMock()
+  mock_response.status_code = 200
+  mock_response.json.return_value = {
+   "choices": [{"message": {"content": "OK"}, "finish_reason": "stop"}]
+  }
+  mock_post.return_value = mock_response
+  
+  ai_client.send(md_content="Context", user_message="Message 1", base_dir=".")
+  
+  args, kwargs = mock_post.call_args
+  payload = kwargs["json"]
+  
+  assert payload["model"] == "deepseek-reasoner"
+  assert "tools" not in payload
+  assert "temperature" not in payload
+  assert "max_tokens" not in payload

tests/test_gui2_layout.py

@@ -19,11 +19,12 @@ def test_gui2_hubs_exist_in_show_windows(app_instance: App) -> None:
 def test_gui2_old_windows_removed_from_show_windows(app_instance: App) -> None:
  """
     Verifies that the old fragmented windows are removed from show_windows.
+    Note: Message, Response, and Tool Calls are kept as they are now optional standalone windows.
     """
  old_windows = [
   "Projects", "Files", "Screenshots",
   "Provider", "System Prompts",
-  "Message", "Response", "Tool Calls", "Comms History"
+  "Comms History"
  ]
  for old_win in old_windows:
   assert old_win not in app_instance.show_windows, f"Old window '{old_win}' should have been removed from show_windows"

tests/test_log_management_ui.py

@@ -77,7 +77,11 @@ def test_render_log_management_logic(app_instance: App) -> None:
  patch("src.gui_2.imgui.table_headers_row"), \
  patch("src.gui_2.imgui.table_next_row"), \
  patch("src.gui_2.imgui.table_next_column"), \
+ patch("src.gui_2.imgui.same_line"), \
+ patch("src.gui_2.imgui.text_colored"), \
+ patch("src.gui_2.imgui.separator"), \
  patch("src.gui_2.imgui.button"):
+
   mock_reg = MockRegistry.return_value
   mock_reg.data = {
    "session_1": {

tests/test_visual_mma.py

@@ -39,11 +39,18 @@ def test_visual_mma_components(live_gui):
  client.push_event("mma_state_update", payload)
  print("  - MMA state update pushed.")
 
- # Use ApiHookClient.wait_for_event if we had a specific event to wait for,
+ # Poll for state update
- # but here we just want to verify state.
+ success = False
- time.sleep(1) # Small sleep for UI to catch up with event queue
+ for _ in range(50): # 10 seconds total
+  if client.get_value("mma_active_tier") == "Tier 3":
+   success = True
+   break
+  time.sleep(0.2)
+
+ assert success, f"State did not update to Tier 3. Current: {client.get_value('mma_active_tier')}"
 
  # 3. Trigger HITL modal
+
  print("Pushing 'mma_step_approval' event to trigger HITL modal...")
  approval_payload = {
   "ticket_id": "TICKET-002",