#!/bin/bash
# Test script for Local LLM Agent Gateway endpoints
# Usage: ./test-endpoints.sh [base_url]

BASE_URL="${1:-http://localhost:3160}"

echo "=============================================="
echo "Testing Local LLM Agent Gateway"
echo "Base URL: $BASE_URL"
echo "=============================================="
echo ""

# Test 1: Health check
echo "1. Testing GET /health"
echo "-------------------------------------------"
curl -s "$BASE_URL/health" | python -m json.tool 2>/dev/null || curl -s "$BASE_URL/health"
echo ""
echo ""

# Test 2: Liveness check
echo "2. Testing GET /health/live"
echo "-------------------------------------------"
curl -s "$BASE_URL/health/live" | python -m json.tool 2>/dev/null || curl -s "$BASE_URL/health/live"
echo ""
echo ""

# Test 3: Readiness check
echo "3. Testing GET /health/ready"
echo "-------------------------------------------"
curl -s "$BASE_URL/health/ready" | python -m json.tool 2>/dev/null || curl -s "$BASE_URL/health/ready"
echo ""
echo ""

# Test 4: List models
echo "4. Testing GET /v1/models"
echo "-------------------------------------------"
curl -s "$BASE_URL/v1/models" | python -m json.tool 2>/dev/null || curl -s "$BASE_URL/v1/models"
echo ""
echo ""

# Test 5: Chat completion (requires Inference Engine running)
echo "5. Testing POST /v1/chat/completions"
echo "-------------------------------------------"
curl -s -X POST "$BASE_URL/v1/chat/completions" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-oss-20b",
    "messages": [
      {"role": "user", "content": "Hello, respond with just \"Hi!\""}
    ],
    "max_tokens": 50,
    "temperature": 0.7
  }' | python -m json.tool 2>/dev/null || echo "Chat completion requires Inference Engine + Ollama running"
echo ""
echo ""

echo "=============================================="
echo "Testing complete!"
echo "=============================================="