chore: C2 DELETE 204+body, C4 cached_tokens, C5 utcnow + LICENSE (T011)

agent_platform/api.py

@@ -8,7 +8,7 @@ from typing import Optional, Dict, Any, List
 from contextlib import asynccontextmanager
 from pathlib import Path
 
-from fastapi import FastAPI, Depends, HTTPException, Request
+from fastapi import FastAPI, Depends, HTTPException, Request, Response
 from fastapi.responses import HTMLResponse, RedirectResponse
 from fastapi.staticfiles import StaticFiles
 from fastapi.templating import Jinja2Templates
@@ -198,10 +198,10 @@ async def api_delete_agent(
         # Code-Agents: nur deaktivieren, damit sie beim nächsten Sync wieder da sind
         await set_agent_enabled(db, agent_id, False)
         await log_audit(db, agent_id=agent_id, user_id=user_id, action="agent_disabled_via_delete", target=agent_id)
-        return {"status": "disabled", "note": "Code-Agents werden nicht gelöscht, nur deaktiviert"}
+        return Response(status_code=204)
     await delete_agent(db, agent_id)
     await log_audit(db, agent_id=agent_id, user_id=user_id, action="agent_deleted", target=agent_id)
-    return {"status": "deleted"}
+    return Response(status_code=204)
 
 
 @app.post("/api/agents/{agent_id}/enable")

agent_platform/db.py

@@ -8,7 +8,7 @@ import aiosqlite
 import json
 import uuid
 import secrets
-from datetime import datetime, timedelta
+from datetime import datetime, timedelta, timezone
 from pathlib import Path
 from config import settings
 
@@ -267,7 +267,7 @@ async def list_audit(db, agent_id=None, limit=100):
 
 async def create_session(db, user_id="default", ttl_hours=24):
     token = secrets.token_urlsafe(32)
-    expires = (datetime.utcnow() + timedelta(hours=ttl_hours)).isoformat()
+    expires = (datetime.now(timezone.utc).replace(tzinfo=None) + timedelta(hours=ttl_hours)).isoformat()
     await db.execute(
         "INSERT INTO sessions (token, user_id, expires_at) VALUES (?, ?, ?)",
         (token, user_id, expires)
@@ -284,6 +284,6 @@ async def validate_session(db, token):
         row = await cur.fetchone()
         if not row:
             return None
-        if row[1] and datetime.fromisoformat(row[1]) < datetime.utcnow():
+        if row[1] and datetime.fromisoformat(row[1]) < datetime.now(timezone.utc).replace(tzinfo=None):
             return None
         return row[0]

agent_platform/llm.py

@@ -72,10 +72,20 @@ async def chat(
 
     usage = {}
     if hasattr(response, "usage") and response.usage:
+        # LiteLLM exposes cached prompt tokens via prompt_tokens_details.cached_tokens
+        # (see litellm.types.utils.Usage). Providers like Anthropic/OpenAI surface this
+        # when prompt caching is active; surfacing it here flows through LLMResponse.usage
+        # to any audit log that serialises the usage dict.
+        cached_tokens = 0
+        details = getattr(response.usage, "prompt_tokens_details", None)
+        if details is not None:
+            cached_tokens = getattr(details, "cached_tokens", None) or 0
+
         usage = {
             "prompt_tokens": response.usage.prompt_tokens or 0,
             "completion_tokens": response.usage.completion_tokens or 0,
             "total_tokens": response.usage.total_tokens or 0,
+            "cached_tokens": cached_tokens,
             "finish_reason": response.choices[0].finish_reason or "stop",
         }