build_push.sh: show current tag, offer to set new one before build

- run_dev.py: starts proxy, admin and frontend together with colored
  output per process; handles SIGTERM from PyCharm Stop button
- .idea/runConfigurations/Dev.xml: PyCharm Python run config
- .gitignore: allow Dev.xml via layered negation pattern
- vite.config.js: open browser automatically on dev server start

.dockerignore

@@ -1,11 +1,50 @@
-.git/
+# Python
+__pycache__/
+*.pyc
+*.pyo
+*.pyd
 .venv/
 venv/
+*.egg-info/
+.pytest_cache/
+
+# Tests
+backend/tests/
+
+# Environment & secrets
 .env
+*.env.local
+
+# Databases
+*.db
+*.sqlite3
+
+# Logs
+logs/
+*.log
+
+# IDE & tools
+.idea/
+.vscode/
+.claude/
+
+# Git
+.git/
+.gitignore
+
+# Frontend
 frontend/node_modules/
 frontend/dist/
-backend/__pycache__/
+
-backend/**/__pycache__/
+# Docker
-backend/*.pyc
+docker-compose.yml
-backend/test.db
+.dockerignore
-backend/tests/
+
+# Docs
+*.md
+
+# Dev & build scripts
+run_dev.py
+build_push.sh
+start.sh
+test_api.sh

.gitignore

@@ -14,7 +14,11 @@ __pycache__/
 *.sqlite3
 
 # IDE
-.idea/
+.idea/*
+!.idea/runConfigurations/
+.idea/runConfigurations/*
+!.idea/runConfigurations/Dev.xml
+!.idea/runConfigurations/Docker_Push.xml
 .vscode/
 
 # Frontend build

.idea/runConfigurations/Dev.xml

@@ -0,0 +1,16 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<component name="ProjectRunConfigurationManager">
+  <configuration default="false" name="Dev" type="PythonConfigurationType" factoryName="Python">
+    <module name="llm_quota" />
+    <option name="INTERPRETER_OPTIONS" value="" />
+    <option name="PARENT_ENVS" value="true" />
+    <option name="SDK_HOME" value="" />
+    <option name="WORKING_DIRECTORY" value="$PROJECT_DIR$" />
+    <option name="IS_MODULE_SDK" value="false" />
+    <option name="ADD_CONTENT_ROOTS" value="true" />
+    <option name="ADD_SOURCE_ROOTS" value="true" />
+    <option name="SCRIPT_NAME" value="$PROJECT_DIR$/run_dev.py" />
+    <option name="PARAMETERS" value="" />
+    <method v="2" />
+  </configuration>
+</component>

.idea/runConfigurations/Docker_Push.xml

@@ -0,0 +1,12 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<component name="ProjectRunConfigurationManager">
+  <configuration default="false" name="Docker Push" type="ShConfigurationType">
+    <option name="SCRIPT_PATH" value="$PROJECT_DIR$/build_push.sh" />
+    <option name="INDEPENDENT_SCRIPT_PATH" value="true" />
+    <option name="INTERPRETER_PATH" value="/bin/bash" />
+    <option name="INTERPRETER_OPTIONS" value="" />
+    <option name="INDEPENDENT_SCRIPT_WORKING_DIRECTORY" value="true" />
+    <option name="SCRIPT_WORKING_DIRECTORY" value="$PROJECT_DIR$" />
+    <method v="2" />
+  </configuration>
+</component>

DOCKERHUB.en.md

@@ -0,0 +1,206 @@
+# mediaeng/llmproxy
+
+A lightweight reverse proxy for [Ollama](https://ollama.com) that manages API keys with configurable token and request quotas. Incoming requests in OpenAI-compatible format are authenticated, checked against the quota, and forwarded to the configured Ollama server.
+
+Ollama does not need to run on the same host — `OLLAMA_URL` can point to any reachable server: the Docker host itself, another machine on the network, or a remote server.
+
+## Features
+
+- OpenAI-compatible endpoint (`/v1/chat/completions`, `/v1/models`)
+- API key management with daily and monthly token/request limits
+- Web-based admin interface (port 8001)
+- Streaming support (Server-Sent Events)
+- Tool use / function calling passthrough
+- Rotating usage logs
+- SQLite (default) or PostgreSQL
+
+## Ports
+
+| Port | Service |
+|------|---------|
+| `8000` | Proxy endpoint (OpenAI API) |
+| `8001` | Admin API + web interface (do not expose) |
+
+## Environment Variables
+
+| Variable | Default | Description |
+|----------|---------|-------------|
+| `ADMIN_PASSWORD` | – | **Required.** Password for the admin interface |
+| `OLLAMA_URL` | `http://localhost:11434` | URL of the Ollama server (without `/v1` suffix) |
+| `DEFAULT_MODEL` | `llama3` | Model used when the client does not specify one |
+| `DATABASE_URL` | `sqlite:///./test.db` | Database connection string (SQLite or PostgreSQL) |
+| `PROXY_HOST` | `0.0.0.0` | Proxy bind address |
+| `PROXY_PORT` | `8000` | Proxy port |
+| `ADMIN_PORT` | `8001` | Admin API port |
+| `APP_TZ` | `Europe/Berlin` | Timezone for daily/monthly quota resets |
+| `LOG_FILE` | `logs/usage.log` | Path of the rotating usage log file |
+
+## Docker Compose – External Ollama, SQLite
+
+Use this when Ollama runs outside of Docker — on the Docker host or any other reachable server. Adjust `OLLAMA_URL` accordingly.
+
+```yaml
+services:
+  llmproxy:
+    image: mediaeng/llmproxy:latest
+    restart: unless-stopped
+    ports:
+      - "8000:8000"
+    environment:
+      ADMIN_PASSWORD: changeme
+      OLLAMA_URL: http://host.docker.internal:11434  # or http://<ip>:11434
+      DEFAULT_MODEL: llama3
+      APP_TZ: Europe/Berlin
+    volumes:
+      - llmproxy-data:/app/backend
+    # On Linux, add extra_hosts since host.docker.internal is not
+    # available automatically:
+    # extra_hosts:
+    #   - "host.docker.internal:host-gateway"
+
+volumes:
+  llmproxy-data:
+```
+
+## Docker Compose – External Ollama, PostgreSQL
+
+```yaml
+services:
+  llmproxy:
+    image: mediaeng/llmproxy:latest
+    restart: unless-stopped
+    ports:
+      - "8000:8000"
+    environment:
+      ADMIN_PASSWORD: changeme
+      OLLAMA_URL: http://host.docker.internal:11434  # or http://<ip>:11434
+      DEFAULT_MODEL: llama3
+      APP_TZ: Europe/Berlin
+      DATABASE_URL: postgresql://llmproxy:secret@db:5432/llmproxy
+    volumes:
+      - llmproxy-data:/app/backend
+    depends_on:
+      db:
+        condition: service_healthy
+    # extra_hosts:
+    #   - "host.docker.internal:host-gateway"
+
+  db:
+    image: postgres:16-alpine
+    restart: unless-stopped
+    environment:
+      POSTGRES_DB: llmproxy
+      POSTGRES_USER: llmproxy
+      POSTGRES_PASSWORD: secret
+    volumes:
+      - pg-data:/var/lib/postgresql/data
+    healthcheck:
+      test: ["CMD-SHELL", "pg_isready -U llmproxy"]
+      interval: 5s
+      timeout: 5s
+      retries: 5
+
+volumes:
+  pg-data:
+```
+
+## Docker Compose – Ollama as Container, SQLite
+
+Ollama and llmproxy run together in Docker, data persisted in a volume.
+
+```yaml
+services:
+  llmproxy:
+    image: mediaeng/llmproxy:latest
+    restart: unless-stopped
+    ports:
+      - "8000:8000"
+    environment:
+      ADMIN_PASSWORD: changeme
+      OLLAMA_URL: http://ollama:11434
+      DEFAULT_MODEL: llama3
+      APP_TZ: Europe/Berlin
+    volumes:
+      - llmproxy-data:/app/backend
+    depends_on:
+      - ollama
+
+  ollama:
+    image: ollama/ollama:latest
+    restart: unless-stopped
+    volumes:
+      - ollama-data:/root/.ollama
+
+volumes:
+  llmproxy-data:
+  ollama-data:
+```
+
+## Docker Compose – Ollama as Container, PostgreSQL
+
+For production environments with an external database.
+
+```yaml
+services:
+  llmproxy:
+    image: mediaeng/llmproxy:latest
+    restart: unless-stopped
+    ports:
+      - "8000:8000"
+    environment:
+      ADMIN_PASSWORD: changeme
+      OLLAMA_URL: http://ollama:11434
+      DEFAULT_MODEL: llama3
+      APP_TZ: Europe/Berlin
+      DATABASE_URL: postgresql://llmproxy:secret@db:5432/llmproxy
+    depends_on:
+      db:
+        condition: service_healthy
+      ollama:
+        condition: service_started
+
+  db:
+    image: postgres:16-alpine
+    restart: unless-stopped
+    environment:
+      POSTGRES_DB: llmproxy
+      POSTGRES_USER: llmproxy
+      POSTGRES_PASSWORD: secret
+    volumes:
+      - pg-data:/var/lib/postgresql/data
+    healthcheck:
+      test: ["CMD-SHELL", "pg_isready -U llmproxy"]
+      interval: 5s
+      timeout: 5s
+      retries: 5
+
+  ollama:
+    image: ollama/ollama:latest
+    restart: unless-stopped
+    volumes:
+      - ollama-data:/root/.ollama
+
+volumes:
+  pg-data:
+  ollama-data:
+```
+
+## Quick Start
+
+```bash
+docker run -d \
+  -p 8000:8000 \
+  -e ADMIN_PASSWORD=changeme \
+  -e OLLAMA_URL=http://host.docker.internal:11434 \
+  -v llmproxy-data:/app/backend \
+  mediaeng/llmproxy:latest
+```
+
+## Client Configuration
+
+Configure the proxy as an OpenAI-compatible endpoint:
+
+```
+Base URL:  http://<host>:8000/v1
+API Key:   <API key created in the admin interface>
+```

DOCKERHUB.md

@@ -0,0 +1,206 @@
+# mediaeng/llmproxy
+
+Ein schlanker Reverse-Proxy für [Ollama](https://ollama.com), der API-Keys mit konfigurierbaren Token- und Request-Quoten verwaltet. Eingehende Anfragen im OpenAI-kompatiblen Format werden authentifiziert, auf Quota geprüft und an den konfigurierten Ollama-Server weitergeleitet.
+
+Ollama muss dabei nicht auf demselben Host laufen — `OLLAMA_URL` kann auf jeden erreichbaren Server zeigen, also auf den Docker-Host selbst, einen anderen Rechner im Netzwerk oder einen Remote-Server.
+
+## Funktionen
+
+- OpenAI-kompatibler Endpunkt (`/v1/chat/completions`, `/v1/models`)
+- API-Key-Verwaltung mit tages- und monatlichen Token-/Request-Limits
+- Web-basierte Admin-Oberfläche (Port 8001)
+- Streaming-Support (Server-Sent Events)
+- Tool-Use / Function Calling wird durchgereicht
+- Rotierende Nutzungs-Logs
+- SQLite (Standard) oder PostgreSQL
+
+## Ports
+
+| Port | Dienst |
+|------|--------|
+| `8000` | Proxy-Endpunkt (OpenAI-API) |
+| `8001` | Admin-API + Web-Oberfläche (nicht exponieren) |
+
+## Umgebungsvariablen
+
+| Variable | Standard | Beschreibung |
+|----------|----------|--------------|
+| `ADMIN_PASSWORD` | – | **Pflicht.** Passwort für die Admin-Oberfläche |
+| `OLLAMA_URL` | `http://localhost:11434` | URL des Ollama-Servers (ohne `/v1`-Suffix) |
+| `DEFAULT_MODEL` | `llama3` | Modell, das verwendet wird wenn der Client keines angibt |
+| `DATABASE_URL` | `sqlite:///./test.db` | Datenbank-Verbindungsstring (SQLite oder PostgreSQL) |
+| `PROXY_HOST` | `0.0.0.0` | Bind-Adresse des Proxy |
+| `PROXY_PORT` | `8000` | Port des Proxy |
+| `ADMIN_PORT` | `8001` | Port der Admin-API |
+| `APP_TZ` | `Europe/Berlin` | Zeitzone für Tages-/Monats-Reset der Quoten |
+| `LOG_FILE` | `logs/usage.log` | Pfad der rotierenden Nutzungs-Logdatei |
+
+## Docker Compose – Ollama extern, SQLite
+
+Wenn Ollama außerhalb von Docker läuft — auf dem Docker-Host oder einem anderen erreichbaren Server. `OLLAMA_URL` entsprechend anpassen.
+
+```yaml
+services:
+  llmproxy:
+    image: mediaeng/llmproxy:latest
+    restart: unless-stopped
+    ports:
+      - "8000:8000"
+    environment:
+      ADMIN_PASSWORD: changeme
+      OLLAMA_URL: http://host.docker.internal:11434  # oder http://<ip>:11434
+      DEFAULT_MODEL: llama3
+      APP_TZ: Europe/Berlin
+    volumes:
+      - llmproxy-data:/app/backend
+    # Auf Linux extra_hosts ergänzen, da host.docker.internal dort
+    # nicht automatisch verfügbar ist:
+    # extra_hosts:
+    #   - "host.docker.internal:host-gateway"
+
+volumes:
+  llmproxy-data:
+```
+
+## Docker Compose – Ollama extern, PostgreSQL
+
+```yaml
+services:
+  llmproxy:
+    image: mediaeng/llmproxy:latest
+    restart: unless-stopped
+    ports:
+      - "8000:8000"
+    environment:
+      ADMIN_PASSWORD: changeme
+      OLLAMA_URL: http://host.docker.internal:11434  # oder http://<ip>:11434
+      DEFAULT_MODEL: llama3
+      APP_TZ: Europe/Berlin
+      DATABASE_URL: postgresql://llmproxy:secret@db:5432/llmproxy
+    volumes:
+      - llmproxy-data:/app/backend
+    depends_on:
+      db:
+        condition: service_healthy
+    # extra_hosts:
+    #   - "host.docker.internal:host-gateway"
+
+  db:
+    image: postgres:16-alpine
+    restart: unless-stopped
+    environment:
+      POSTGRES_DB: llmproxy
+      POSTGRES_USER: llmproxy
+      POSTGRES_PASSWORD: secret
+    volumes:
+      - pg-data:/var/lib/postgresql/data
+    healthcheck:
+      test: ["CMD-SHELL", "pg_isready -U llmproxy"]
+      interval: 5s
+      timeout: 5s
+      retries: 5
+
+volumes:
+  pg-data:
+```
+
+## Docker Compose – Ollama als Container, SQLite
+
+Ollama und llmproxy laufen gemeinsam in Docker, Daten in einem Volume.
+
+```yaml
+services:
+  llmproxy:
+    image: mediaeng/llmproxy:latest
+    restart: unless-stopped
+    ports:
+      - "8000:8000"
+    environment:
+      ADMIN_PASSWORD: changeme
+      OLLAMA_URL: http://ollama:11434
+      DEFAULT_MODEL: llama3
+      APP_TZ: Europe/Berlin
+    volumes:
+      - llmproxy-data:/app/backend
+    depends_on:
+      - ollama
+
+  ollama:
+    image: ollama/ollama:latest
+    restart: unless-stopped
+    volumes:
+      - ollama-data:/root/.ollama
+
+volumes:
+  llmproxy-data:
+  ollama-data:
+```
+
+## Docker Compose – Ollama als Container, PostgreSQL
+
+Für Produktionsumgebungen mit externer Datenbank.
+
+```yaml
+services:
+  llmproxy:
+    image: mediaeng/llmproxy:latest
+    restart: unless-stopped
+    ports:
+      - "8000:8000"
+    environment:
+      ADMIN_PASSWORD: changeme
+      OLLAMA_URL: http://ollama:11434
+      DEFAULT_MODEL: llama3
+      APP_TZ: Europe/Berlin
+      DATABASE_URL: postgresql://llmproxy:secret@db:5432/llmproxy
+    depends_on:
+      db:
+        condition: service_healthy
+      ollama:
+        condition: service_started
+
+  db:
+    image: postgres:16-alpine
+    restart: unless-stopped
+    environment:
+      POSTGRES_DB: llmproxy
+      POSTGRES_USER: llmproxy
+      POSTGRES_PASSWORD: secret
+    volumes:
+      - pg-data:/var/lib/postgresql/data
+    healthcheck:
+      test: ["CMD-SHELL", "pg_isready -U llmproxy"]
+      interval: 5s
+      timeout: 5s
+      retries: 5
+
+  ollama:
+    image: ollama/ollama:latest
+    restart: unless-stopped
+    volumes:
+      - ollama-data:/root/.ollama
+
+volumes:
+  pg-data:
+  ollama-data:
+```
+
+## Schnellstart
+
+```bash
+docker run -d \
+  -p 8000:8000 \
+  -e ADMIN_PASSWORD=changeme \
+  -e OLLAMA_URL=http://host.docker.internal:11434 \
+  -v llmproxy-data:/app/backend \
+  mediaeng/llmproxy:latest
+```
+
+## Client-Konfiguration
+
+Den Proxy als OpenAI-kompatibler Endpunkt konfigurieren:
+
+```
+Base URL:  http://<host>:8000/v1
+API Key:   <angelegter API-Key aus der Admin-Oberfläche>
+```

Dockerfile

@@ -9,7 +9,7 @@ FROM python:3.12-slim
 WORKDIR /app
 
 COPY backend/requirements.txt .
-RUN pip install --no-cache-dir -r requirements.txt
+RUN pip install --no-cache-dir -r requirements.txt psycopg2-binary
 
 COPY backend/ backend/
 COPY --from=frontend-builder /app/frontend/dist frontend/dist

README.md

@@ -85,7 +85,7 @@ docker build -t llm-quota .
 ```bash
 docker run -d \
   -p 8000:8000 \
-  -p 8001:8001 \
+  -p 127.0.0.1:8001:8001 \
   -e ADMIN_PASSWORD=geheim \
   -e OLLAMA_URL=http://host.docker.internal:11434 \
   -e DATABASE_URL=sqlite:///./data/quota.db \
@@ -94,27 +94,39 @@ docker run -d \
   llm-quota
 ```
 
-| Port | Dienst |
+| Port | Bindung | Dienst |
-|------|--------|
+|------|---------|--------|
-| `8000` | Proxy (für LLM-Clients) |
+| `8000` | `0.0.0.0` — öffentlich | Proxy (für LLM-Clients) |
-| `8001` | Admin-API + Admin-Oberfläche |
+| `8001` | `127.0.0.1` — nur lokal am Server | Admin-API + Admin-Oberfläche |
 
-Admin-Oberfläche: `http://localhost:8001`
+Docker unterscheidet beim Port-Mapping zwischen `0.0.0.0` (alle Interfaces, öffentlich erreichbar) und `127.0.0.1` (nur der Server selbst kann zugreifen). Mit `-p 127.0.0.1:8001:8001` ist Port 8001 am Server verfügbar, aber von außen nicht direkt ansprechbar.
+
+### Admin-Oberfläche per SSH-Tunnel erreichbar machen
+
+Der SSH-Tunnel leitet einen lokalen Port auf den Server weiter und nutzt dabei, dass Port 8001 dort auf `127.0.0.1` erreichbar ist:
+
+```
+Admin-Laptop:8001  ──SSH──►  Server:127.0.0.1:8001  ──►  Container:8001
+```
+
+```bash
+ssh -L 8001:localhost:8001 user@server
+```
+
+Danach ist die Admin-Oberfläche auf dem Laptop unter `http://localhost:8001` erreichbar — ohne dass Port 8001 öffentlich exponiert wird.
 
 ### Mit PostgreSQL
 
 ```bash
 docker run -d \
   -p 8000:8000 \
-  -p 8001:8001 \
+  -p 127.0.0.1:8001:8001 \
   -e ADMIN_PASSWORD=geheim \
   -e DATABASE_URL=postgresql://user:pass@db-host:5432/llm_quota \
   -e OLLAMA_URL=http://ollama:11434 \
   llm-quota
 ```
 
-> **Hinweis:** Im Container bindet die Admin-API auf `0.0.0.0`. Port 8001 sollte nicht öffentlich exponiert werden — entweder per Firewall absichern oder hinter einem Reverse-Proxy (nginx, Caddy) betreiben.
-
 ## Proxy-Endpunkte (Port 8000)
 
 Alle Endpunkte erfordern einen gültigen API-Key im `Authorization`-Header.

backend/admin.py

@@ -8,7 +8,7 @@ from fastapi.staticfiles import StaticFiles
 from sqlalchemy.orm import Session
 from database import get_db
 import crud, schemas
-from models import APIKey as APIKeyModel
+from models import APIKey as APIKeyModel, Usage as UsageModel
 
 app = FastAPI(title="Ollama Proxy Admin API")
 
@@ -31,13 +31,24 @@ def require_admin_auth(request: Request):
     if not secrets.compare_digest(token, ADMIN_PASSWORD):
         raise HTTPException(status_code=401, detail="Invalid admin password")
 
-@app.get("/api/api-keys", response_model=list[schemas.APIKey])
+@app.get("/api/api-keys", response_model=list[schemas.APIKeyWithUsage])
 async def read_api_keys(
     skip: int = 0, limit: int = 100,
     db: Session = Depends(get_db),
     _ = Depends(require_admin_auth),
 ):
-    return db.query(APIKeyModel).offset(skip).limit(limit).all()
+    keys = db.query(APIKeyModel).offset(skip).limit(limit).all()
+    result = []
+    for key in keys:
+        item = schemas.APIKeyWithUsage.model_validate(key)
+        usage = db.query(UsageModel).filter(UsageModel.api_key_id == key.id).first()
+        if usage:
+            item.tokens_used_today = usage.tokens_used_today or 0
+            item.tokens_used_month = usage.tokens_used_month or 0
+            item.requests_today = usage.requests_today or 0
+            item.requests_month = usage.requests_month or 0
+        result.append(item)
+    return result
 
 @app.post("/api/api-keys", response_model=schemas.APIKeyCreated)
 async def create_api_key(
@@ -87,6 +98,32 @@ async def deactivate_api_key(
     db.commit()
     return {"message": "API key deactivated"}
 
+@app.put("/api/api-keys/{api_key_id}/activate")
+async def activate_api_key(
+    api_key_id: int,
+    db: Session = Depends(get_db),
+    _ = Depends(require_admin_auth),
+):
+    db_key = db.query(APIKeyModel).filter(APIKeyModel.id == api_key_id).first()
+    if not db_key:
+        raise HTTPException(status_code=404, detail="API key not found")
+    db_key.is_active = True
+    db.commit()
+    return {"message": "API key activated"}
+
+@app.delete("/api/api-keys/{api_key_id}")
+async def delete_api_key(
+    api_key_id: int,
+    db: Session = Depends(get_db),
+    _ = Depends(require_admin_auth),
+):
+    db_key = db.query(APIKeyModel).filter(APIKeyModel.id == api_key_id).first()
+    if not db_key:
+        raise HTTPException(status_code=404, detail="API key not found")
+    db.delete(db_key)
+    db.commit()
+    return {"message": "API key deleted"}
+
 @app.get("/api/proxy-info")
 async def get_proxy_info(_ = Depends(require_admin_auth)):
     host = os.getenv("PROXY_HOST", "0.0.0.0")
@@ -107,13 +144,18 @@ async def update_settings(
     db: Session = Depends(get_db),
     _ = Depends(require_admin_auth),
 ):
-    crud.set_setting(db, "ollama_url", settings.ollama_url)
+    ollama_url = settings.ollama_url.rstrip('/').removesuffix('/v1')
+    crud.set_setting(db, "ollama_url", ollama_url)
     crud.set_setting(db, "default_model", settings.default_model)
-    return settings
+    return schemas.Settings(ollama_url=ollama_url, default_model=settings.default_model)
 
 @app.get("/api/ollama-models")
-async def get_ollama_models(db: Session = Depends(get_db), _ = Depends(require_admin_auth)):
+async def get_ollama_models(
-    ollama_url = crud.get_setting(db, "ollama_url", "http://localhost:11434")
+    url: str = None,
+    db: Session = Depends(get_db),
+    _ = Depends(require_admin_auth),
+):
+    ollama_url = url or crud.get_setting(db, "ollama_url", "http://localhost:11434")
     try:
         async with httpx.AsyncClient(timeout=5.0) as client:
             response = await client.get(f"{ollama_url}/api/tags")

backend/crud.py

@@ -40,6 +40,7 @@ def create_api_key(
     db_key = APIKey(
         name=name,
         key=_hash_api_key(raw_key),
+        key_prefix=raw_key[:12],
         expires_at=expires_at,
         daily_tokens=daily_tokens,
         monthly_tokens=monthly_tokens,

backend/main.py

@@ -1,135 +1,133 @@
-import time
+import logging
-import uuid
+import os
+from logging.handlers import RotatingFileHandler
+from pathlib import Path
+
 from fastapi import FastAPI, HTTPException, Depends, Request
-from fastapi.responses import JSONResponse
+from fastapi.responses import JSONResponse, StreamingResponse
 from sqlalchemy.orm import Session
-from database import get_db, SessionLocal
+from database import get_db
 import crud
 import httpx
-import os
 
-app = FastAPI(title="Ollama Proxy")
+# Rotating usage log (8 KB per file, 3 backups)
+_log_path = Path(os.getenv("LOG_FILE", "logs/usage.log"))
+_log_path.parent.mkdir(parents=True, exist_ok=True)
+_handler = RotatingFileHandler(str(_log_path), maxBytes=8192, backupCount=3, encoding="utf-8")
+_handler.setFormatter(logging.Formatter("%(asctime)s | %(message)s", datefmt="%Y-%m-%d %H:%M:%S"))
+usage_log = logging.getLogger("proxy.usage")
+usage_log.setLevel(logging.INFO)
+usage_log.addHandler(_handler)
+usage_log.propagate = False
 
-async def proxy_request(url: str, method: str = "GET", json_data: dict = None, headers: dict = None):
+def _last_user_msg(messages: list, max_len: int = 120) -> str:
-    async with httpx.AsyncClient() as client:
+    for msg in reversed(messages):
-        response = await client.request(method=method, url=url, json=json_data, headers=headers)
+        if msg.get("role") == "user":
-        return response
+            text = (msg.get("content") or "").replace("\n", " ").strip()
+            return text[:max_len] + ("…" if len(text) > max_len else "")
+    return ""
 
-@app.middleware("http")
+async def require_api_key(request: Request, db: Session = Depends(get_db)):
-async def authenticate_and_quota(request: Request, call_next):
     auth_header = request.headers.get("Authorization", "")
     if auth_header.startswith("Bearer "):
-        api_key = auth_header.replace("Bearer ", "")
+        api_key = auth_header[7:]
     elif auth_header.startswith("sk-"):
         api_key = auth_header
     else:
-        return JSONResponse(status_code=401, content={"detail": "Invalid or missing API key"})
+        raise HTTPException(status_code=401, detail="Invalid or missing API key")
+    db_key = crud.verify_api_key(db, api_key)
+    if not db_key:
+        raise HTTPException(status_code=401, detail="Invalid API key")
+    request.state.api_key_id = db_key.id
+    request.state.api_key_name = db_key.name
 
-    # Uses its own session since middleware cannot use Depends
+app = FastAPI(title="Ollama Proxy", dependencies=[Depends(require_api_key)])
-    db = SessionLocal()
-    try:
-        db_key = crud.verify_api_key(db, api_key)
-        if not db_key:
-            return JSONResponse(status_code=401, content={"detail": "Invalid API key"})
-        request.state.api_key_id = db_key.id
-    finally:
-        db.close()
 
-    response = await call_next(request)
+async def proxy_request(url: str, method: str = "GET", json_data: dict = None):
-    return response
+    async with httpx.AsyncClient(timeout=300.0) as client:
+        response = await client.request(method=method, url=url, json=json_data)
+        return response
 
 @app.post("/api/generate")
 async def generate(request: Request, db: Session = Depends(get_db)):
-    api_key_id = request.state.api_key_id
     ollama_url = crud.get_setting(db, "ollama_url", os.getenv("OLLAMA_URL", "http://localhost:11434"))
-
     body = await request.json()
-
     prompt_tokens = crud.count_tokens(body.get("prompt", ""))
-    if not crud.check_and_increment_quota(db, api_key_id, tokens=prompt_tokens, requests=1):
+
+    if not crud.check_and_increment_quota(db, request.state.api_key_id, tokens=prompt_tokens, requests=1):
         raise HTTPException(status_code=429, detail="Quota exceeded")
 
-    response = await proxy_request(f"{ollama_url}/api/generate", method="POST", json_data=body, headers=dict(request.headers))
+    prompt_preview = (body.get("prompt", "").replace("\n", " ").strip())[:120]
-    return JSONResponse(content=response.json(), status_code=response.status_code, headers=dict(response.headers))
+    usage_log.info('%s | /api/generate | %s | ~%d tokens | "%s"',
+                   request.state.api_key_name, body.get("model", "?"), prompt_tokens, prompt_preview)
+    response = await proxy_request(f"{ollama_url}/api/generate", method="POST", json_data=body)
+    return JSONResponse(content=response.json(), status_code=response.status_code)
 
 @app.post("/api/chat")
 async def chat(request: Request, db: Session = Depends(get_db)):
-    api_key_id = request.state.api_key_id
     ollama_url = crud.get_setting(db, "ollama_url", os.getenv("OLLAMA_URL", "http://localhost:11434"))
-
     body = await request.json()
+    messages = body.get("messages", [])
+    prompt_tokens = sum(crud.count_tokens(msg.get("content") or "") for msg in messages)
 
-    prompt_tokens = sum(crud.count_tokens(msg.get("content", "")) for msg in body.get("messages", []))
+    if not crud.check_and_increment_quota(db, request.state.api_key_id, tokens=prompt_tokens, requests=1):
-    if not crud.check_and_increment_quota(db, api_key_id, tokens=prompt_tokens, requests=1):
         raise HTTPException(status_code=429, detail="Quota exceeded")
 
-    response = await proxy_request(f"{ollama_url}/api/chat", method="POST", json_data=body, headers=dict(request.headers))
+    usage_log.info('%s | /api/chat | %s | ~%d tokens | "%s"',
-    return JSONResponse(content=response.json(), status_code=response.status_code, headers=dict(response.headers))
+                   request.state.api_key_name, body.get("model", "?"), prompt_tokens, _last_user_msg(messages))
+    response = await proxy_request(f"{ollama_url}/api/chat", method="POST", json_data=body)
+    return JSONResponse(content=response.json(), status_code=response.status_code)
 
 @app.get("/api/tags")
-async def list_models(request: Request, db: Session = Depends(get_db)):
+async def list_models(db: Session = Depends(get_db)):
     ollama_url = crud.get_setting(db, "ollama_url", os.getenv("OLLAMA_URL", "http://localhost:11434"))
-    response = await proxy_request(f"{ollama_url}/api/tags", method="GET", headers=dict(request.headers))
+    response = await proxy_request(f"{ollama_url}/api/tags", method="GET")
-    return JSONResponse(content=response.json(), status_code=response.status_code, headers=dict(response.headers))
+    return JSONResponse(content=response.json(), status_code=response.status_code)
 
 @app.get("/api/versions")
-async def versions(request: Request, db: Session = Depends(get_db)):
+async def versions(db: Session = Depends(get_db)):
     ollama_url = crud.get_setting(db, "ollama_url", os.getenv("OLLAMA_URL", "http://localhost:11434"))
-    response = await proxy_request(f"{ollama_url}/api/versions", method="GET", headers=dict(request.headers))
+    response = await proxy_request(f"{ollama_url}/api/versions", method="GET")
-    return JSONResponse(content=response.json(), status_code=response.status_code, headers=dict(response.headers))
+    return JSONResponse(content=response.json(), status_code=response.status_code)
 
 @app.get("/v1/models")
-async def list_openai_models(request: Request, db: Session = Depends(get_db)):
+async def list_openai_models(db: Session = Depends(get_db)):
     ollama_url = crud.get_setting(db, "ollama_url", os.getenv("OLLAMA_URL", "http://localhost:11434"))
-    response = await proxy_request(f"{ollama_url}/api/tags", method="GET", headers=dict(request.headers))
+    response = await proxy_request(f"{ollama_url}/v1/models", method="GET")
-    ollama_models = response.json()
+    return JSONResponse(content=response.json(), status_code=response.status_code)
-    openai_models = {
-        "object": "list",
-        "data": [
-            {
-                "id": model["name"],
-                "object": "model",
-                "created": int(model["modified_at"][:10].replace("-", "")) * 1000 if "modified_at" in model else 0,
-                "owned_by": "ollama"
-            }
-            for model in ollama_models.get("models", [])
-        ]
-    }
-    return JSONResponse(content=openai_models, status_code=200, headers=dict(response.headers))
 
 @app.post("/v1/chat/completions")
 async def openai_chat_completions(request: Request, db: Session = Depends(get_db)):
-    api_key_id = request.state.api_key_id
     ollama_url = crud.get_setting(db, "ollama_url", os.getenv("OLLAMA_URL", "http://localhost:11434"))
     default_model = crud.get_setting(db, "default_model", os.getenv("DEFAULT_MODEL", "llama3"))
 
     body = await request.json()
     messages = body.get("messages", [])
-    prompt_tokens = sum(crud.count_tokens(msg.get("content", "")) for msg in messages)
+    prompt_tokens = sum(crud.count_tokens(msg.get("content") or "") for msg in messages)
 
-    if not crud.check_and_increment_quota(db, api_key_id, tokens=prompt_tokens, requests=1):
+    if not crud.check_and_increment_quota(db, request.state.api_key_id, tokens=prompt_tokens, requests=1):
         raise HTTPException(status_code=429, detail="Quota exceeded")
 
-    ollama_body = {
+    if "model" not in body:
-        "model": body.get("model", default_model),
+        body = {**body, "model": default_model}
-        "messages": messages,
-        "stream": body.get("stream", False)
-    }
 
-    response = await proxy_request(f"{ollama_url}/api/chat", method="POST", json_data=ollama_body, headers=dict(request.headers))
+    model_name = body["model"]
-    response_content = response.json().get("message", {}).get("content", "")
+    usage_log.info('%s | /v1/chat/completions | %s | ~%d tokens | "%s"',
-    completion_tokens = crud.count_tokens(response_content)
+                   request.state.api_key_name, model_name, prompt_tokens, _last_user_msg(messages))
 
-    openai_response = {
+    target = f"{ollama_url}/v1/chat/completions"
-        "id": f"chatcmpl-{uuid.uuid4().hex}",
+
-        "object": "chat.completion",
+    if body.get("stream"):
-        "created": int(time.time()),
+        async def generate():
-        "model": body.get("model", default_model),
+            async with httpx.AsyncClient(timeout=300.0) as client:
-        "choices": [{"index": 0, "message": {"role": "assistant", "content": response_content}, "finish_reason": "stop"}],
+                async with client.stream("POST", target, json=body) as resp:
-        "usage": {
+                    async for chunk in resp.aiter_bytes():
-            "prompt_tokens": prompt_tokens,
+                        yield chunk
-            "completion_tokens": completion_tokens,
+        return StreamingResponse(
-            "total_tokens": prompt_tokens + completion_tokens,
+            generate(),
-        },
+            media_type="text/event-stream",
-    }
+            headers={"Cache-Control": "no-cache", "X-Accel-Buffering": "no"},
-    return JSONResponse(content=openai_response, status_code=200, headers={"Content-Type": "application/json"})
+        )
+
+    response = await proxy_request(target, method="POST", json_data=body)
+    return JSONResponse(content=response.json(), status_code=response.status_code)

backend/models.py

@@ -10,6 +10,7 @@ class APIKey(Base):
     id = Column(Integer, primary_key=True, index=True)
     name = Column(String)
     key = Column(String, unique=True, index=True)
+    key_prefix = Column(String)
     is_active = Column(Boolean, default=True)
     created_at = Column(DateTime(timezone=True), default=_now)
     expires_at = Column(DateTime(timezone=True), nullable=True)

backend/schemas.py

@@ -14,6 +14,7 @@ class APIKey(BaseModel):
     id: int
     name: str
     key: str
+    key_prefix: Optional[str] = None
     is_active: bool
     created_at: datetime
     expires_at: Optional[datetime] = None
@@ -51,3 +52,12 @@ class UsageStats(BaseModel):
 
     class Config:
         from_attributes = True
+
+class APIKeyWithUsage(APIKey):
+    tokens_used_today: int = 0
+    tokens_used_month: int = 0
+    requests_today: int = 0
+    requests_month: int = 0
+
+    class Config:
+        from_attributes = True

build_push.sh

@@ -0,0 +1,40 @@
+#!/bin/bash
+set -e
+cd "$(dirname "$0")"
+
+IMAGE=mediaeng/llmproxy
+PLATFORM=linux/arm64
+
+CURRENT=$(git describe --tags --always)
+if [ -z "$CURRENT" ]; then
+    echo "Fehler: git describe liefert kein Ergebnis"
+    exit 1
+fi
+
+echo "Aktueller Tag: $CURRENT"
+read -rp "Neuer Tag [${CURRENT}]: " INPUT
+VERSION="${INPUT:-$CURRENT}"
+
+if [ "$VERSION" != "$CURRENT" ]; then
+    git tag "$VERSION"
+    git push origin "$VERSION"
+    echo "Tag '$VERSION' gesetzt und gepusht."
+fi
+
+echo ""
+echo "Image   : $IMAGE"
+echo "Platform: $PLATFORM"
+echo "Tags    : $IMAGE:$VERSION  $IMAGE:latest"
+echo ""
+
+docker buildx build \
+    --platform "$PLATFORM" \
+    --push \
+    -t "$IMAGE:$VERSION" \
+    -t "$IMAGE:latest" \
+    .
+
+echo ""
+echo "Fertig. Gepusht:"
+echo "  docker pull $IMAGE:$VERSION"
+echo "  docker pull $IMAGE:latest"

docker-compose.yml

@@ -1,24 +1,10 @@
-backend:
+services:
-  build:
+  llmproxy:
-    context: ./backend
+    image: mediaeng/llmproxy:latest
-    dockerfile: Dockerfile
+    restart: unless-stopped
-  ports:
+    env_file: .env
-    - "8000:8000"
+    ports:
-  environment:
+      - "${PROXY_PORT:-8000}:${PROXY_PORT:-8000}"
-    - DATABASE_URL=postgresql://ollama:password@db:5432/ollama_proxy
+    volumes:
-    - OLLAMA_URL=http://ollama:11434
+      - ./backend/test.db:/app/backend/test.db
-    - SECRET_KEY=your-secret-key-change-me
+      - ./backend/logs:/app/backend/logs
-  depends_on:
-    - db
-
-db:
-  image: postgres:16
-  environment:
-    - POSTGRES_USER=ollama
-    - POSTGRES_PASSWORD=password
-    - POSTGRES_DB=ollama_proxy
-  volumes:
-    - postgres_data:/var/lib/postgresql/data
-
-volumes:
-  postgres_data:

frontend/src/main.jsx

@@ -3,12 +3,29 @@ import ReactDOM from 'react-dom/client';
 import axios from 'axios';
 import './styles.css';
 
-const maskKey = (key) => `••••••••${key.slice(-4)}`;
+const displayKey = (prefix) => prefix ? `${prefix}••••••••` : '••••••••••••';
 
 function authHeaders(token) {
   return { Authorization: `Bearer ${token}` };
 }
 
+const fmtK = (n) => { const k = n / 1000; return k % 1 === 0 ? `${k}k` : `${k.toFixed(1)}k`; };
+
+function QuotaBar({ used, limit, isToken = false }) {
+  if (limit == null) return <span className="quota-unlimited">∞</span>;
+  const pct = Math.min(100, (used / limit) * 100);
+  const color = pct >= 90 ? '#e74c3c' : pct >= 70 ? '#e67e22' : '#27ae60';
+  const fmt = isToken ? fmtK : (n) => n.toLocaleString('de-DE');
+  return (
+    <div className="quota-cell">
+      <span className="quota-label">{fmt(used)} / {fmt(limit)}</span>
+      <div className="progress-bar">
+        <div className="progress-fill" style={{ width: `${pct}%`, backgroundColor: color }} />
+      </div>
+    </div>
+  );
+}
+
 function Login({ onLogin }) {
   const [password, setPassword] = useState('');
   const [error, setError] = useState(null);
@@ -51,20 +68,40 @@ const EMPTY_KEY_FORM = {
 function SettingsSection({ password }) {
   const [settings, setSettings] = useState(null);
   const [availableModels, setAvailableModels] = useState([]);
+  const [modelsLoading, setModelsLoading] = useState(false);
   const [proxyEndpoint, setProxyEndpoint] = useState(null);
   const [saved, setSaved] = useState(false);
   const [error, setError] = useState(null);
 
+  const fetchModels = async (url, currentModel) => {
+    setModelsLoading(true);
+    try {
+      const res = await axios.get('/api/ollama-models', {
+        headers: authHeaders(password),
+        params: url ? { url } : {},
+      });
+      const models = res.data.models;
+      setAvailableModels(models);
+      if (models.length > 0 && !models.includes(currentModel)) {
+        setSettings(s => ({ ...s, default_model: models[0] }));
+      }
+    } catch {
+      setAvailableModels([]);
+    } finally {
+      setModelsLoading(false);
+    }
+  };
+
   useEffect(() => {
     const headers = authHeaders(password);
     Promise.all([
       axios.get('/api/settings', { headers }),
-      axios.get('/api/ollama-models', { headers }),
       axios.get('/api/proxy-info', { headers }),
-    ]).then(([settingsRes, modelsRes, proxyRes]) => {
+    ]).then(([settingsRes, proxyRes]) => {
-      setSettings(settingsRes.data);
+      const s = settingsRes.data;
-      setAvailableModels(modelsRes.data.models);
+      setSettings(s);
       setProxyEndpoint(proxyRes.data.endpoint);
+      fetchModels(s.ollama_url, s.default_model);
     }).catch(() => setError('Einstellungen konnten nicht geladen werden.'));
   }, []);
 
@@ -100,13 +137,16 @@ function SettingsSection({ password }) {
             type="url"
             value={settings.ollama_url}
             onChange={(e) => setSettings({ ...settings, ollama_url: e.target.value })}
+            onBlur={(e) => fetchModels(e.target.value, settings.default_model)}
             placeholder="http://localhost:11434"
             required
           />
         </div>
         <div className="settings-row">
           <label>Standard-Modell</label>
-          {availableModels.length > 0 ? (
+          {modelsLoading ? (
+            <span className="settings-value">Lade Modelle…</span>
+          ) : availableModels.length > 0 ? (
             <select
               value={settings.default_model}
               onChange={(e) => setSettings({ ...settings, default_model: e.target.value })}
@@ -139,6 +179,8 @@ function App() {
   const [newKey, setNewKey] = useState(null);
   const [form, setForm] = useState(EMPTY_KEY_FORM);
   const [creating, setCreating] = useState(false);
+  const [editKey, setEditKey] = useState(null);
+  const [editForm, setEditForm] = useState({});
 
   useEffect(() => {
     if (!password) { setLoading(false); return; }
@@ -160,8 +202,8 @@ function App() {
     try {
       const payload = { name: form.name };
       if (form.expires_at)       payload.expires_at       = new Date(form.expires_at).toISOString();
-      if (form.daily_tokens)     payload.daily_tokens     = Number(form.daily_tokens);
+      if (form.daily_tokens)     payload.daily_tokens     = Number(form.daily_tokens) * 1000;
-      if (form.monthly_tokens)   payload.monthly_tokens   = Number(form.monthly_tokens);
+      if (form.monthly_tokens)   payload.monthly_tokens   = Number(form.monthly_tokens) * 1000;
       if (form.daily_requests)   payload.daily_requests   = Number(form.daily_requests);
       if (form.monthly_requests) payload.monthly_requests = Number(form.monthly_requests);
 
@@ -185,6 +227,53 @@ function App() {
     }
   };
 
+  const handleActivate = async (id) => {
+    try {
+      await axios.put(`/api/api-keys/${id}/activate`, {}, { headers: authHeaders(password) });
+      await fetchApiKeys();
+    } catch {
+      setError('Fehler beim Aktivieren.');
+    }
+  };
+
+  const handleDelete = async (id, name) => {
+    if (!window.confirm(`API-Key "${name}" wirklich löschen?`)) return;
+    try {
+      await axios.delete(`/api/api-keys/${id}`, { headers: authHeaders(password) });
+      if (editKey?.id === id) setEditKey(null);
+      await fetchApiKeys();
+    } catch {
+      setError('Fehler beim Löschen.');
+    }
+  };
+
+  const handleEdit = (key) => {
+    setEditKey(key);
+    setEditForm({
+      daily_tokens:     key.daily_tokens    != null ? key.daily_tokens / 1000    : '',
+      monthly_tokens:   key.monthly_tokens  != null ? key.monthly_tokens / 1000  : '',
+      daily_requests:   key.daily_requests  ?? '',
+      monthly_requests: key.monthly_requests ?? '',
+    });
+  };
+
+  const handleSaveEdit = async (e) => {
+    e.preventDefault();
+    try {
+      const payload = {
+        daily_tokens:     editForm.daily_tokens     !== '' ? Number(editForm.daily_tokens) * 1000     : null,
+        monthly_tokens:   editForm.monthly_tokens   !== '' ? Number(editForm.monthly_tokens) * 1000   : null,
+        daily_requests:   editForm.daily_requests   !== '' ? Number(editForm.daily_requests)   : null,
+        monthly_requests: editForm.monthly_requests !== '' ? Number(editForm.monthly_requests) : null,
+      };
+      await axios.patch(`/api/api-keys/${editKey.id}/quota`, payload, { headers: authHeaders(password) });
+      setEditKey(null);
+      await fetchApiKeys();
+    } catch {
+      setError('Fehler beim Speichern der Limits.');
+    }
+  };
+
   const logout = () => {
     sessionStorage.removeItem('admin_password');
     setPassword(null);
@@ -206,43 +295,76 @@ function App() {
       <section>
         <h2>Neuer API-Key</h2>
         <form onSubmit={handleCreate} className="create-form">
-          <input
+          <div className="edit-form">
-            placeholder="Name"
+            <label className="create-name">
-            value={form.name}
+              Name
-            onChange={(e) => setForm({ ...form, name: e.target.value })}
+              <small>&nbsp;</small>
-            required
+              <input
-          />
+                value={form.name}
-          <input
+                onChange={(e) => setForm({ ...form, name: e.target.value })}
-            type="date"
+                required
-            placeholder="Ablaufdatum (leer = unbegrenzt)"
+                placeholder="z. B. alice"
-            value={form.expires_at}
+              />
-            onChange={(e) => setForm({ ...form, expires_at: e.target.value })}
+            </label>
-          />
+            <label className="create-date">
-          <input
+              Ablaufdatum
-            type="number"
+              <small>leer = unbegrenzt</small>
-            placeholder="Tokens/Tag (leer = unbegrenzt)"
+              <input
-            value={form.daily_tokens}
+                type="date"
-            onChange={(e) => setForm({ ...form, daily_tokens: e.target.value })}
+                value={form.expires_at}
-          />
+                onChange={(e) => setForm({ ...form, expires_at: e.target.value })}
-          <input
+              />
-            type="number"
+            </label>
-            placeholder="Tokens/Monat"
+            <div className="create-btn-wrap">
-            value={form.monthly_tokens}
+              <button type="submit" disabled={creating} className="btn-save">Erstellen</button>
-            onChange={(e) => setForm({ ...form, monthly_tokens: e.target.value })}
+            </div>
-          />
+          </div>
-          <input
+          <div className="edit-form">
-            type="number"
+            <label>
-            placeholder="Requests/Tag"
+              Tokens/Tag (k)
-            value={form.daily_requests}
+              <small>leer = unbegrenzt</small>
-            onChange={(e) => setForm({ ...form, daily_requests: e.target.value })}
+              <input
-          />
+                type="number"
-          <input
+                min="0"
-            type="number"
+                placeholder="∞"
-            placeholder="Requests/Monat"
+                value={form.daily_tokens}
-            value={form.monthly_requests}
+                onChange={(e) => setForm({ ...form, daily_tokens: e.target.value })}
-            onChange={(e) => setForm({ ...form, monthly_requests: e.target.value })}
+              />
-          />
+            </label>
-          <button type="submit" disabled={creating}>Erstellen</button>
+            <label>
+              Tokens/Monat (k)
+              <small>leer = unbegrenzt</small>
+              <input
+                type="number"
+                min="0"
+                placeholder="∞"
+                value={form.monthly_tokens}
+                onChange={(e) => setForm({ ...form, monthly_tokens: e.target.value })}
+              />
+            </label>
+            <label>
+              Requests/Tag
+              <small>leer = unbegrenzt</small>
+              <input
+                type="number"
+                min="0"
+                placeholder="∞"
+                value={form.daily_requests}
+                onChange={(e) => setForm({ ...form, daily_requests: e.target.value })}
+              />
+            </label>
+            <label>
+              Requests/Monat
+              <small>leer = unbegrenzt</small>
+              <input
+                type="number"
+                min="0"
+                placeholder="∞"
+                value={form.monthly_requests}
+                onChange={(e) => setForm({ ...form, monthly_requests: e.target.value })}
+              />
+            </label>
+          </div>
         </form>
         {newKey && (
           <div className="new-key-box">
@@ -258,41 +380,99 @@ function App() {
         <table>
           <thead>
             <tr>
-              <th>ID</th>
+              <th></th>
               <th>Name</th>
               <th>Key</th>
-              <th>Status</th>
               <th>Läuft ab</th>
-              <th>Tokens/Tag</th>
+              <th className="th-quota">Tokens/Tag</th>
-              <th>Tokens/Monat</th>
+              <th className="th-quota">Tokens/Monat</th>
-              <th>Req/Tag</th>
+              <th className="th-quota">Req/Tag</th>
-              <th>Req/Monat</th>
+              <th className="th-quota">Req/Monat</th>
               <th></th>
             </tr>
           </thead>
           <tbody>
             {apiKeys.map(key => (
-              <tr key={key.id}>
+              <tr key={key.id} className={editKey?.id === key.id ? 'row-editing' : ''}>
-                <td>{key.id}</td>
+                <td className="td-status">
+                  <span className={key.is_active ? 'status-active' : 'status-inactive'} title={key.is_active ? 'Aktiv' : 'Inaktiv'}>●</span>
+                </td>
                 <td>{key.name}</td>
-                <td>{maskKey(key.key)}</td>
+                <td>{displayKey(key.key_prefix)}</td>
-                <td>{key.is_active ? 'Aktiv' : 'Inaktiv'}</td>
                 <td>{key.expires_at ? new Date(key.expires_at).toLocaleDateString('de-DE', { timeZone: 'Europe/Berlin' }) : '∞'}</td>
-                <td>{key.daily_tokens ?? '∞'}</td>
+                <td><QuotaBar used={key.tokens_used_today} limit={key.daily_tokens} isToken /></td>
-                <td>{key.monthly_tokens ?? '∞'}</td>
+                <td><QuotaBar used={key.tokens_used_month} limit={key.monthly_tokens} isToken /></td>
-                <td>{key.daily_requests ?? '∞'}</td>
+                <td><QuotaBar used={key.requests_today} limit={key.daily_requests} /></td>
-                <td>{key.monthly_requests ?? '∞'}</td>
+                <td><QuotaBar used={key.requests_month} limit={key.monthly_requests} /></td>
-                <td>
+                <td className="action-cell">
-                  {key.is_active && (
+                  <button className="btn-icon btn-icon-edit" data-tooltip="Bearbeiten" onClick={() => handleEdit(key)}>✏</button>
-                    <button className="btn-danger" onClick={() => handleDeactivate(key.id)}>
+                  {key.is_active ? (
-                      Deaktivieren
+                    <button className="btn-icon btn-icon-warn" data-tooltip="Deaktivieren" onClick={() => handleDeactivate(key.id)}>⏸</button>
-                    </button>
+                  ) : (
+                    <button className="btn-icon btn-icon-ok" data-tooltip="Aktivieren" onClick={() => handleActivate(key.id)}>▶</button>
                   )}
+                  <button className="btn-icon btn-icon-danger" data-tooltip="Löschen" onClick={() => handleDelete(key.id, key.name)}>✕</button>
                 </td>
               </tr>
             ))}
           </tbody>
         </table>
+
+        {editKey && (
+          <div className="edit-section">
+            <h3>Limits bearbeiten: <em>{editKey.name}</em></h3>
+            <form onSubmit={handleSaveEdit} className="edit-form">
+              <label>
+                Tokens/Tag (k)
+                <small>leer = unbegrenzt</small>
+                <input
+                  type="number"
+                  min="0"
+                  value={editForm.daily_tokens}
+                  onChange={(e) => setEditForm({ ...editForm, daily_tokens: e.target.value })}
+                  placeholder="∞"
+                />
+              </label>
+              <label>
+                Tokens/Monat (k)
+                <small>leer = unbegrenzt</small>
+                <input
+                  type="number"
+                  min="0"
+                  value={editForm.monthly_tokens}
+                  onChange={(e) => setEditForm({ ...editForm, monthly_tokens: e.target.value })}
+                  placeholder="∞"
+                />
+              </label>
+              <label>
+                Requests/Tag
+                <small>leer = unbegrenzt</small>
+                <input
+                  type="number"
+                  min="0"
+                  value={editForm.daily_requests}
+                  onChange={(e) => setEditForm({ ...editForm, daily_requests: e.target.value })}
+                  placeholder="∞"
+                />
+              </label>
+              <label>
+                Requests/Monat
+                <small>leer = unbegrenzt</small>
+                <input
+                  type="number"
+                  min="0"
+                  value={editForm.monthly_requests}
+                  onChange={(e) => setEditForm({ ...editForm, monthly_requests: e.target.value })}
+                  placeholder="∞"
+                />
+              </label>
+              <div className="edit-actions">
+                <button type="submit" className="btn-save">Speichern</button>
+                <button type="button" className="btn-cancel" onClick={() => setEditKey(null)}>Abbrechen</button>
+              </div>
+            </form>
+          </div>
+        )}
       </section>
     </div>
   );

frontend/src/styles.css

@@ -92,30 +92,37 @@ tr:hover {
 
 .create-form {
   display: flex;
-  flex-wrap: wrap;
+  flex-direction: column;
-  gap: 10px;
+  gap: 12px;
+}
+
+.create-form .edit-form {
+  margin-bottom: 0;
+}
+
+.create-name {
+  flex: 0 0 auto !important;
+}
+
+.create-name input {
+  width: 200px;
+}
+
+.create-date {
+  flex: 0 0 auto !important;
+}
+
+.create-date input[type="date"] {
+  width: 150px;
+}
+
+.create-btn-wrap {
+  flex: 0 0 auto;
+  display: flex;
   align-items: flex-end;
 }
 
-.create-form input {
+.create-form .btn-save:disabled {
-  padding: 8px 10px;
-  border: 1px solid #ccc;
-  border-radius: 4px;
-  font-size: 14px;
-  flex: 1 1 160px;
-}
-
-.create-form button {
-  padding: 8px 20px;
-  background: #27ae60;
-  color: white;
-  border: none;
-  border-radius: 4px;
-  cursor: pointer;
-  font-size: 14px;
-}
-
-.create-form button:disabled {
   opacity: 0.6;
   cursor: not-allowed;
 }
@@ -218,3 +225,198 @@ tr:hover {
   color: #999;
   font-size: 12px;
 }
+
+/* Quota progress bars */
+.th-quota {
+  width: 90px;
+}
+
+.quota-cell {
+  width: 100%;
+}
+
+.quota-label {
+  font-size: 12px;
+  color: #555;
+  white-space: nowrap;
+}
+
+.quota-unlimited {
+  color: #aaa;
+  font-size: 14px;
+}
+
+.progress-bar {
+  height: 4px;
+  background: #e2e8f0;
+  border-radius: 2px;
+  margin-top: 4px;
+}
+
+.progress-fill {
+  height: 100%;
+  border-radius: 2px;
+  transition: width 0.3s ease;
+}
+
+/* Edit form */
+.td-status {
+  width: 20px;
+  text-align: center;
+}
+
+.status-active  { color: #27ae60; font-size: 12px; }
+.status-inactive { color: #bdc3c7; font-size: 12px; }
+
+.btn-icon {
+  background: none;
+  border: none;
+  cursor: pointer;
+  font-size: 16px;
+  padding: 2px 4px;
+  margin-right: 2px;
+  border-radius: 3px;
+  line-height: 1;
+  position: relative;
+}
+
+.btn-icon:hover {
+  background: rgba(0, 0, 0, 0.06);
+}
+
+.btn-icon-edit   { color: #2980b9; }
+.btn-icon-warn   { color: #e67e22; }
+.btn-icon-ok     { color: #27ae60; }
+.btn-icon-danger { color: #e74c3c; }
+
+.btn-icon::after {
+  content: attr(data-tooltip);
+  position: absolute;
+  bottom: calc(100% + 8px);
+  left: 50%;
+  transform: translateX(-50%);
+  background: #2c3e50;
+  color: white;
+  padding: 5px 10px;
+  border-radius: 5px;
+  font-size: 12px;
+  font-weight: 400;
+  white-space: nowrap;
+  opacity: 0;
+  pointer-events: none;
+  transition: opacity 0.15s;
+  z-index: 1000;
+}
+
+.btn-icon::before {
+  content: '';
+  position: absolute;
+  bottom: calc(100% + 2px);
+  left: 50%;
+  transform: translateX(-50%);
+  border: 5px solid transparent;
+  border-top-color: #2c3e50;
+  opacity: 0;
+  pointer-events: none;
+  transition: opacity 0.15s;
+  z-index: 1000;
+}
+
+.btn-icon:hover::after,
+.btn-icon:hover::before {
+  opacity: 1;
+}
+
+.action-cell {
+  white-space: nowrap;
+  position: relative;
+  overflow: visible;
+}
+
+.row-editing {
+  background: #eaf4fb !important;
+}
+
+.edit-section {
+  margin-top: 20px;
+  padding: 16px 20px;
+  background: #f0f7ff;
+  border: 1px solid #3498db;
+  border-radius: 6px;
+}
+
+.edit-section h3 {
+  margin: 0 0 14px;
+  font-size: 15px;
+  color: #2c3e50;
+  font-weight: 600;
+}
+
+.edit-section h3 em {
+  font-style: normal;
+  color: #2980b9;
+}
+
+.edit-form {
+  display: flex;
+  flex-wrap: wrap;
+  gap: 14px;
+  align-items: flex-end;
+}
+
+.edit-form label {
+  display: flex;
+  flex-direction: column;
+  gap: 3px;
+  font-size: 13px;
+  font-weight: 500;
+  color: #34495e;
+}
+
+.edit-form label small {
+  font-weight: 400;
+  color: #999;
+  font-size: 11px;
+}
+
+.edit-form input {
+  padding: 6px 8px;
+  border: 1px solid #bdc3c7;
+  border-radius: 4px;
+  font-size: 14px;
+  width: 130px;
+}
+
+.edit-actions {
+  display: flex;
+  gap: 8px;
+  padding-bottom: 2px;
+}
+
+.btn-save {
+  padding: 6px 18px;
+  background: #27ae60;
+  color: white;
+  border: none;
+  border-radius: 4px;
+  cursor: pointer;
+  font-size: 13px;
+}
+
+.btn-save:hover {
+  background: #219a52;
+}
+
+.btn-cancel {
+  padding: 6px 18px;
+  background: #95a5a6;
+  color: white;
+  border: none;
+  border-radius: 4px;
+  cursor: pointer;
+  font-size: 13px;
+}
+
+.btn-cancel:hover {
+  background: #7f8c8d;
+}

run_dev.py

@@ -0,0 +1,86 @@
+#!/usr/bin/env python3
+"""Dev runner: startet Proxy, Admin-API und Frontend gemeinsam."""
+import os
+import sys
+import signal
+import subprocess
+import threading
+from pathlib import Path
+
+
+def load_env(path: Path):
+    if not path.exists():
+        return
+    for line in path.read_text(encoding='utf-8').splitlines():
+        line = line.strip()
+        if line and not line.startswith('#') and '=' in line:
+            key, _, val = line.partition('=')
+            os.environ.setdefault(key.strip(), val.strip().strip('"\''))
+
+
+def pipe_output(proc: subprocess.Popen, label: str, color: str):
+    reset = '\033[0m'
+    prefix = f'\033[{color}m[{label}]{reset} '
+    for raw in iter(proc.stdout.readline, b''):
+        sys.stdout.write(prefix + raw.decode(errors='replace'))
+        sys.stdout.flush()
+
+
+def main():
+    root = Path(__file__).parent
+    load_env(root / '.env')
+
+    if not os.environ.get('ADMIN_PASSWORD'):
+        print('Fehler: ADMIN_PASSWORD nicht gesetzt (siehe .env)')
+        sys.exit(1)
+
+    python  = root / '.venv' / 'bin' / 'python'
+    backend = root / 'backend'
+    frontend = root / 'frontend'
+
+    proxy_host = os.environ.get('PROXY_HOST', '0.0.0.0')
+    proxy_port = os.environ.get('PROXY_PORT', '8000')
+    admin_port = os.environ.get('ADMIN_PORT', '8001')
+
+    print('Initialisiere Datenbank...')
+    subprocess.run([str(python), 'init_db.py'], cwd=backend, check=True)
+
+    print(f'Starte Proxy      → http://{proxy_host}:{proxy_port}')
+    print(f'Starte Admin-API  → http://127.0.0.1:{admin_port}')
+    print('Starte Frontend   → http://localhost:5173')
+
+    env = {**os.environ, 'PYTHONUNBUFFERED': '1'}
+
+    procs = [
+        (subprocess.Popen(
+            [str(python), '-m', 'uvicorn', 'main:app', '--reload',
+             '--host', proxy_host, '--port', proxy_port],
+            cwd=backend, stdout=subprocess.PIPE, stderr=subprocess.STDOUT, env=env,
+        ), 'Proxy   ', '34'),  # blau
+        (subprocess.Popen(
+            [str(python), '-m', 'uvicorn', 'admin:app', '--reload',
+             '--host', '127.0.0.1', '--port', admin_port],
+            cwd=backend, stdout=subprocess.PIPE, stderr=subprocess.STDOUT, env=env,
+        ), 'Admin   ', '33'),  # gelb
+        (subprocess.Popen(
+            ['npm', 'run', 'dev'],
+            cwd=frontend, stdout=subprocess.PIPE, stderr=subprocess.STDOUT,
+        ), 'Frontend', '32'),  # grün
+    ]
+
+    for proc, label, color in procs:
+        threading.Thread(target=pipe_output, args=(proc, label, color), daemon=True).start()
+
+    def stop(*_):
+        for p, _, _ in procs:
+            p.terminate()
+
+    signal.signal(signal.SIGINT, stop)
+    signal.signal(signal.SIGTERM, stop)
+
+    for p, _, _ in procs:
+        p.wait()
+
+
+if __name__ == '__main__':
+    main()