#!/bin/bash
# model-usage-tracker.sh — Check model availability and report usage
# Runs daily to warn if models are nearing credit limits

PRIMARY_MODEL="deepseek-chat"
PRIMARY_PROVIDER="admin-ai"

echo "=== Model Usage Check $(date +'%b %d %Y %I:%M %p') ==="

# Test DeepSeek (primary)
RESULT=$(curl -s -X POST "https://admin-ai.itpropartner.com/v1/chat/completions" \
  -H "Authorization: Bearer $(grep -A2 'admin-ai:' ~/.hermes/config.yaml | grep api_key | awk '{print $2}')" \
  -H "Content-Type: application/json" \
  -d "{\"model\":\"$PRIMARY_MODEL\",\"messages\":[{\"role\":\"user\",\"content\":\"ping\"}],\"max_tokens\":5}" 2>&1)

if echo "$RESULT" | grep -qi "insufficient\|quota\|429\|rate\|exceeded\|credit"; then
    echo "❌ PRIMARY: deepseek-chat — CREDITS EXHAUSTED or RATE LIMITED"
    echo "   Error: $(echo $RESULT | grep -oP '"message":"[^"]*"' | head -1)"
elif echo "$RESULT" | grep -qi "error"; then
    echo "⚠️  PRIMARY: deepseek-chat — Error: $(echo $RESULT | grep -oP '"message":"[^"]*"')"
else
    TOKENS=$(echo "$RESULT" | python3 -c "import sys,json;d=json.load(sys.stdin);u=d['usage'];print(f'prompt={u[\"prompt_tokens\"]} completion={u[\"completion_tokens\"]} total={u[\"total_tokens\"]}')" 2>/dev/null || echo "unknown")
    echo "✅ PRIMARY: deepseek-chat — OK ($TOKENS)"
fi

# Test local fallback
LOCAL=$(curl -s http://localhost:11434/api/generate -d '{"model":"llama3.2:3b","prompt":"ping","stream":false}' 2>&1)
if echo "$LOCAL" | grep -qi "response"; then
    echo "✅ LOCAL: llama3.2:3b — OK"
else
    echo "❌ LOCAL: llama3.2:3b — NOT RESPONDING"
fi