Backend: Python-Logging statt print, Diagnose in allen Fallbacks

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

backend/agents.py

@@ -5,15 +5,19 @@ jeweilige Provider fehl — der andere läuft unverändert weiter.
 """
 
 import asyncio
+import logging
 import os
 import re
 import shutil
 import tempfile
+import time
 import urllib.request
 from pathlib import Path
 
 from config import PROVIDERS, DEFAULT_PROVIDER
 
+log = logging.getLogger("creator.agents")
+
 _active_processes: dict[str, asyncio.subprocess.Process] = {}
 
 # Capability → Claude --allowedTools
@@ -54,7 +58,11 @@ def provider_available(provider: str) -> bool:
 def kill_process(agent_key_prefix: str) -> None:
     """Killt alle aktiven Prozesse, deren Key mit dem Prefix beginnt (deckt -plan/-w1… ab)."""
     for key, process in list(_active_processes.items()):
-        if key.startswith(agent_key_prefix) and process.returncode is None:
+        if process.returncode is not None:  # tote Einträge beim Iterieren aufräumen
+            _active_processes.pop(key, None)
+            continue
+        if key.startswith(agent_key_prefix):
+            log.debug("kill agent %s", key)
             process.kill()
 
 
@@ -76,6 +84,7 @@ async def run_agent(
 
 
 async def _communicate(agent_key: str, cmd: list[str], stdin_data: bytes | None, timeout: int) -> tuple[int, str, str]:
+    start = time.monotonic()
     process = await asyncio.create_subprocess_exec(
         *cmd,
         stdin=asyncio.subprocess.PIPE if stdin_data is not None else asyncio.subprocess.DEVNULL,
@@ -95,10 +104,18 @@ async def _communicate(agent_key: str, cmd: list[str], stdin_data: bytes | None,
                 await asyncio.wait_for(process.wait(), timeout=5)
             except asyncio.TimeoutError:
                 pass
+            log.info("agent %s: Timeout nach %ds", agent_key, timeout)
             raise
+        log.info(
+            "agent %s: exit %s nach %.1fs (%d Bytes stdout)",
+            agent_key, process.returncode, time.monotonic() - start, len(stdout),
+        )
         return process.returncode, stdout.decode("utf-8", errors="replace"), stderr.decode("utf-8", errors="replace")
     finally:
-        _active_processes.pop(agent_key, None)
+        # Pop nur bei Identität: ein Slot-Restart unter demselben Key darf den
+        # NEUEN Prozess nicht aus dem Tracking werfen.
+        if _active_processes.get(agent_key) is process:
+            del _active_processes[agent_key]
 
 
 async def _run_claude_cli(agent_key: str, prompt: str, timeout: int, role: str, capabilities: str) -> tuple[int, str, str]: