Release v2.0.0

repository

  - Change HACS badge from Default to Custom
  - Add custom repository installation steps
  - Update installation instructions
  - Revise documentation format

README.md

@@ -1,73 +1,81 @@
 # 🤖 HA Text AI for Home Assistant
 
-<div align="center">
+<div align="center">  
 
-![GitHub release](https://img.shields.io/github/release/smkrv/ha-text-ai.svg?style=flat-square)
-![GitHub downloads](https://img.shields.io/github/downloads/smkrv/ha-text-ai/total.svg?style=flat-square)
-![GitHub stars](https://img.shields.io/github/stars/smkrv/ha-text-ai.svg?style=social)
-![GitHub last commit](https://img.shields.io/github/last-commit/smkrv/ha-text-ai.svg?style=flat-square)
-[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg?style=flat-square)](https://opensource.org/licenses/MIT)
-[![hacs_badge](https://img.shields.io/badge/HACS-Default-orange.svg?style=flat-square)](https://github.com/hacs/integration)
-[![Community Forum](https://img.shields.io/badge/Community-Forum-blue.svg?style=flat-square)](https://community.home-assistant.io/t/ha-text-ai-integration)
+  ![GitHub release](https://img.shields.io/github/release/smkrv/ha-text-ai.svg?style=flat-square) ![GitHub downloads](https://img.shields.io/github/downloads/smkrv/ha-text-ai/total.svg?style=flat-square) ![GitHub stars](https://img.shields.io/github/stars/smkrv/ha-text-ai.svg?style=social) ![GitHub last commit](https://img.shields.io/github/last-commit/smkrv/ha-text-ai.svg?style=flat-square) [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg?style=flat-square)](https://opensource.org/licenses/MIT) [![hacs_badge](https://img.shields.io/badge/HACS-Custom-41BDF5.svg?style=flat-square)](https://github.com/hacs/integration)
 
 </div>
 
 <p align="center">
-Transform your smart home experience with powerful AI assistance powered by OpenAI's GPT models. Get intelligent responses, automate complex scenarios, and enhance your home automation with natural language processing.
+Transform your smart home experience with powerful AI assistance powered by multiple AI providers including OpenAI GPT and Anthropic Claude models. Get intelligent responses, automate complex scenarios, and enhance your home automation with advanced natural language processing.
 </p>
 
 ---
 
 ## 🌟 Features
 
-- 🧠 **Advanced AI Integration**:
-  - Support for latest GPT models
+- 🧠 **Multi-Provider AI Integration**:
+  - Support for OpenAI GPT models
+  - Anthropic Claude integration
+  - Custom API endpoints
+  - Flexible model selection
+- 💬 **Advanced Language Processing**:
   - Context-aware responses
   - Multi-turn conversations
-- 💬 **Natural Language Control**:
-  - Control devices using everyday language
-  - Get detailed explanations and recommendations
+  - Custom system instructions
   - Natural conversation flow
-- 📝 **Smart Memory Management**:
+- 📝 **Enhanced Memory Management**:
   - Persistent conversation history
   - Context-aware responses
   - Customizable history limits
-- ⚡ **Performance Optimized**:
+  - Model-specific filtering
+- ⚡ **Performance Optimization**:
   - Efficient token usage
-  - Rate limit handling
+  - Smart rate limiting
   - Response caching
+  - Request interval control
 - 🎯 **Advanced Customization**:
-  - Adjustable response parameters
+  - Per-request model selection
+  - Adjustable parameters
   - Custom system prompts
-  - Model selection per request
+  - Temperature control
 - 🔒 **Enhanced Security**:
   - Secure API key storage
   - Rate limiting protection
   - Error handling
-- 🎨 **User Experience**:
+  - Usage monitoring
+- 🎨 **Improved User Experience**:
   - Intuitive configuration UI
   - Detailed sensor attributes
   - Rich service interface
+  - Model selection UI
 - 🔄 **Automation Integration**:
   - Event-driven responses
   - Conditional logic support
   - Template compatibility
+  - Model-specific automation
 
 ## 📋 Prerequisites
 
 - Home Assistant 2023.8.0 or newer
-- OpenAI API key ([Get one here](https://platform.openai.com/account/api-keys))
+- API key from supported providers:
+  - OpenAI ([Get key](https://platform.openai.com/account/api-keys))
+  - Anthropic ([Get key](https://console.anthropic.com/))
 - Python 3.9 or newer
 - Stable internet connection
 
 ## ⚡ Installation
 
-### HACS Installation (Recommended)
-1. Open HACS in Home Assistant
-2. Click the "+" button
-3. Search for "HA Text AI"
-4. Click "Install"
-5. Restart Home Assistant
+### HACS Installation (Recommended)  
+1. Open HACS in Home Assistant  
+2. Click on the three dots in the top right corner  
+3. Select "Custom repositories"  
+4. Add `https://github.com/smkrv/ha-text-ai` as Integration  
+5. Click "Add"  
+6. Click on "Integrations" in HACS  
+7. Search for "HA Text AI"  
+8. Click "Download"  
+9. Restart Home Assistant  
 
 ### Manual Installation
 1. Download the latest release
@@ -86,12 +94,15 @@ Transform your smart home experience with powerful AI assistance powered by Open
 ### Via YAML
 ```yaml
 ha_text_ai:
-  api_key: !secret openai_api_key
-  model: gpt-3.5-turbo
+  api_key: !secret ai_api_key
+  model: gpt-3.5-turbo  # or claude-3-sonnet
   temperature: 0.7
   max_tokens: 1000
   request_interval: 1.0
-  api_endpoint: https://api.openai.com/v1  # optional
+  api_endpoint: https://api.openai.com/v1  # optional, for custom endpoints
+  system_prompt: |
+    You are a home automation expert assistant.
+    Focus on practical and efficient solutions.
 ```
 
 ## 🛠️ Available Services
@@ -101,9 +112,10 @@ ha_text_ai:
 service: ha_text_ai.ask_question
 data:
   question: "What's the optimal temperature for sleeping?"
-  model: "gpt-4o"  # optional
+  model: "claude-3-sonnet"  # optional
   temperature: 0.5  # optional
   max_tokens: 500  # optional
+  system_prompt: "You are a sleep optimization expert"  # optional
 ```
 
 ### set_system_prompt
@@ -128,114 +140,37 @@ service: ha_text_ai.clear_history
 service: ha_text_ai.get_history
 data:
   limit: 5  # optional
+  filter_model: "gpt-4"  # optional
 ```
 
-## 🔧 Advanced Examples
-
-### Smart Energy Management
-```yaml
-automation:
-  alias: "AI Energy Optimization"
-  trigger:
-    platform: time_pattern
-    hours: "/2"
-  action:
-    - service: ha_text_ai.ask_question
-      data:
-        question: >
-          Current power usage: {{ states('sensor.total_power') }}W
-          Temperature: {{ states('sensor.indoor_temperature') }}°C
-          Time: {{ now().strftime('%H:%M') }}
-          Occupancy: {{ states('binary_sensor.occupancy') }}
-
-          Analyze current energy usage and suggest optimizations
-          considering comfort and efficiency.
-        temperature: 0.3
-        max_tokens: 200
-    - service: notify.mobile_app
-      data:
-        message: "{{ states.sensor.ha_text_ai.attributes.response }}"
-```
-
-### Contextual Lighting Control
-```yaml
-automation:
-  alias: "AI Lighting Assistant"
-  trigger:
-    platform: state
-    entity_id: binary_sensor.motion
-  variables:
-    context: >
-      Time: {{ now().strftime('%H:%M') }}
-      Light Level: {{ states('sensor.illuminance') }}
-      Room: {{ trigger.to_state.attributes.room }}
-      Activity: {{ states('input_select.current_activity') }}
-      Weather: {{ states('weather.home') }}
-  action:
-    - service: ha_text_ai.ask_question
-      data:
-        question: >
-          Based on this context:
-          {{ context }}
-
-          Suggest optimal lighting settings for current conditions.
-        model: gpt-3.5-turbo
-        temperature: 0.4
-    - service: scene.turn_on
-      data:
-        entity_id: >
-          {{ states.sensor.ha_text_ai.attributes.response | regex_findall('scene\.[a-z_]+') | first }}
-```
-
-## 📊 Performance Optimization
-
-### Token Usage
-- Use focused system prompts
-- Implement response caching
-- Clear history periodically
-- Monitor token usage
-
-### Response Time
-- Adjust request_interval
-- Use faster models for simple queries
-- Implement timeout handling
-- Cache frequent responses
-
-### Memory Management
-- Set appropriate history limits
-- Clear unused contexts
-- Monitor memory usage
-- Use efficient data structures
-
-## ❗ Troubleshooting
-
-### API Issues
-- Verify API key validity
-- Check rate limits
-- Monitor usage quotas
-- Test endpoint accessibility
-
-### Performance Issues
-- Reduce max_tokens
-- Increase request_interval
-- Clear conversation history
-- Check network connectivity
-
-### Integration Issues
-- Verify HA version compatibility
-- Check component dependencies
-- Review log files
-- Update configuration
-
 ## 📘 FAQ
 
+**Q: Which AI providers are supported?**
+
+A: Currently OpenAI (GPT models) and Anthropic (Claude models) are supported, with more providers planned.
+
 **Q: How can I reduce API costs?**
+
+A: Use GPT-3.5-Turbo or Claude-3-Sonnet for most queries, implement caching, and optimize token usage.
+
+**Q: Can I use custom models?**
+
+A: Yes, you can configure custom endpoints and use any compatible model by specifying it in the configuration.
+
+**Q: How do I switch between different AI providers?**
+
+A: Simply change the model parameter in your configuration or service calls to use the desired provider's model.
+
+**Q: How can I reduce API costs?**
+
 A: Use GPT-3.5-Turbo for most queries, implement caching, and optimize token usage.
 
 **Q: Is my data secure?**
+
 A: Yes, API keys are stored securely and data is transmitted via encrypted connections.
 
 **Q: Can I use custom models?**
+
 A: Yes, configure custom endpoints and models via configuration options.
 
 ## 🤝 Contributing

custom_components/ha_text_ai/__init__.py

@@ -1,14 +1,20 @@
 """The HA Text AI integration."""
 import logging
-from typing import Any
+from typing import Any, Dict, Optional
+import asyncio
+import voluptuous as vol
+import json
+from datetime import datetime
 
 from homeassistant.config_entries import ConfigEntry
 from homeassistant.const import CONF_API_KEY
-from homeassistant.core import HomeAssistant
-from homeassistant.exceptions import ConfigEntryNotReady
+from homeassistant.core import HomeAssistant, ServiceCall, callback
+from homeassistant.exceptions import ConfigEntryNotReady, HomeAssistantError
 from homeassistant.helpers import aiohttp_client
 from homeassistant.helpers import config_validation as cv
+from async_timeout import timeout
 
+from .coordinator import HATextAICoordinator
 from .const import (
     DOMAIN,
     PLATFORMS,
@@ -22,15 +28,94 @@ from .const import (
     DEFAULT_MAX_TOKENS,
     DEFAULT_API_ENDPOINT,
     DEFAULT_REQUEST_INTERVAL,
+    API_VERSION,
+    API_MODELS_PATH,
+    API_CHAT_PATH,
+    API_TIMEOUT,
+    API_RETRY_COUNT,
+    API_BACKOFF_FACTOR,
+    LOGGER_NAME,
+    STATE_ERROR,
+    STATE_READY,
+    STATE_PROCESSING,
+    STATE_RATE_LIMITED,
+    STATE_MAINTENANCE,
+    STATE_DISCONNECTED,
+    STATE_RETRYING,
+    STATE_QUEUED,
+    STATE_UPDATING,
+    SUPPORTED_MODELS,
+    EVENT_RESPONSE_RECEIVED,
+    EVENT_ERROR_OCCURRED,
+    EVENT_STATE_CHANGED,
 )
 
-_LOGGER = logging.getLogger(__name__)
+_LOGGER = logging.getLogger(LOGGER_NAME)
 
 CONFIG_SCHEMA = cv.config_entry_only_config_schema(DOMAIN)
 
+# Service validation schemas
+SERVICE_SCHEMA_ASK_QUESTION = vol.Schema({
+    vol.Required("question"): cv.string,
+    vol.Optional("system_prompt"): cv.string,
+    vol.Optional("model"): vol.In(SUPPORTED_MODELS),
+    vol.Optional("temperature"): vol.All(
+        vol.Coerce(float), vol.Range(min=0, max=2)
+    ),
+    vol.Optional("max_tokens"): vol.All(
+        vol.Coerce(int), vol.Range(min=1, max=4096)
+    ),
+    vol.Optional("priority"): vol.Boolean,
+})
+
+SERVICE_SCHEMA_GET_HISTORY = vol.Schema({
+    vol.Optional("limit", default=10): vol.All(
+        vol.Coerce(int), vol.Range(min=1, max=100)
+    ),
+    vol.Optional("filter_model"): vol.In(SUPPORTED_MODELS),
+    vol.Optional("start_date"): cv.datetime,
+    vol.Optional("include_metadata"): vol.Boolean,
+})
+
+SERVICE_SCHEMA_SET_SYSTEM_PROMPT = vol.Schema({
+    vol.Required("prompt"): cv.string,
+})
+
+async def async_check_api(session, endpoint: str, headers: dict, is_anthropic: bool = False) -> bool:
+    """Check API availability for different providers."""
+    try:
+        if is_anthropic:
+            check_url = f"{endpoint}/v1/models"
+        else:
+            check_url = f"{endpoint}/{API_VERSION}/{API_MODELS_PATH}"
+
+        async with timeout(API_TIMEOUT):
+            async with session.get(check_url, headers=headers) as response:
+                if response.status == 200:
+                    return True
+                elif response.status == 401:
+                    raise ConfigEntryNotReady("Invalid API key")
+                elif response.status == 429:
+                    _LOGGER.warning("Rate limit exceeded during API check")
+                    return False
+                else:
+                    _LOGGER.error("API check failed with status: %d", response.status)
+                    return False
+    except Exception as ex:
+        _LOGGER.error("API check error: %s", str(ex))
+        return False
+
 async def async_setup(hass: HomeAssistant, config: dict[str, Any]) -> bool:
     """Set up the HA Text AI component."""
-    hass.data.setdefault(DOMAIN, {})
+    hass.data.setdefault(DOMAIN, {
+        "coordinators": {},
+        "metrics": {
+            "total_requests": 0,
+            "total_tokens": 0,
+            "errors": {},
+            "model_usage": {},
+        }
+    })
     return True
 
 async def async_setup_entry(hass: HomeAssistant, entry: ConfigEntry) -> bool:
@@ -38,34 +123,198 @@ async def async_setup_entry(hass: HomeAssistant, entry: ConfigEntry) -> bool:
     try:
         session = aiohttp_client.async_get_clientsession(hass)
 
+        # Determine API type based on model
+        model = entry.data.get(CONF_MODEL, DEFAULT_MODEL)
+        is_anthropic = any(model.startswith("claude") for model in SUPPORTED_MODELS)
+
+        api_key = entry.data[CONF_API_KEY]
+        endpoint = entry.data.get(CONF_API_ENDPOINT, DEFAULT_API_ENDPOINT).rstrip('/')
+
+        # Configure headers based on API type
+        headers = {
+            "Content-Type": "application/json",
+            "Accept": "application/json"
+        }
+
+        if is_anthropic:
+            headers["x-api-key"] = api_key
+            headers["anthropic-version"] = "2023-06-01"
+        else:
+            headers["Authorization"] = f"Bearer {api_key}"
+
+        # Check API with retries
+        for attempt in range(API_RETRY_COUNT):
+            if await async_check_api(session, endpoint, headers, is_anthropic):
+                break
+            if attempt < API_RETRY_COUNT - 1:
+                delay = API_BACKOFF_FACTOR * (2 ** attempt)
+                await asyncio.sleep(delay)
+            else:
+                raise ConfigEntryNotReady("Failed to connect to API")
+
+        # Create and initialize coordinator
         coordinator = HATextAICoordinator(
             hass,
-            api_key=entry.data[CONF_API_KEY],
-            endpoint=entry.data.get(CONF_API_ENDPOINT, DEFAULT_API_ENDPOINT),
-            model=entry.data.get(CONF_MODEL, DEFAULT_MODEL),
+            api_key=api_key,
+            endpoint=endpoint,
+            model=model,
             temperature=entry.data.get(CONF_TEMPERATURE, DEFAULT_TEMPERATURE),
             max_tokens=entry.data.get(CONF_MAX_TOKENS, DEFAULT_MAX_TOKENS),
             request_interval=entry.data.get(CONF_REQUEST_INTERVAL, DEFAULT_REQUEST_INTERVAL),
             session=session,
+            is_anthropic=is_anthropic
         )
 
-        try:
-            await coordinator.async_config_entry_first_refresh()
-        except Exception as refresh_ex:
-            _LOGGER.error("Failed to refresh coordinator: %s", str(refresh_ex))
-            return False
+        # Initialize coordinator
+        await coordinator.async_config_entry_first_refresh()
 
-        if not coordinator.last_update_success:
-            _LOGGER.error("Failed to communicate with OpenAI API")
-            return False
+        # Check coordinator status
+        if coordinator.endpoint_status == "auth_error":
+            raise ConfigEntryNotReady("Authentication failed")
+        elif coordinator.endpoint_status == "rate_limited":
+            _LOGGER.warning("API rate limited during setup")
+        elif coordinator.endpoint_status == "maintenance":
+            raise ConfigEntryNotReady("API is in maintenance mode")
+        elif coordinator.endpoint_status == "error":
+            raise ConfigEntryNotReady("API error during setup")
+        elif not coordinator.last_update_success:
+            raise ConfigEntryNotReady("Failed to initialize coordinator")
 
         hass.data[DOMAIN][entry.entry_id] = coordinator
 
-        try:
-            await hass.config_entries.async_forward_entry_setups(entry, PLATFORMS)
-        except Exception as setup_ex:
-            _LOGGER.error("Failed to setup platforms: %s", str(setup_ex))
-            return False
+        await hass.config_entries.async_forward_entry_setups(entry, PLATFORMS)
+
+        # Register event handlers
+        @callback
+        def handle_state_change(event):
+            """Handle state changes."""
+            if event.data.get("entity_id").startswith(f"{DOMAIN}."):
+                _LOGGER.debug("State changed: %s", event.data)
+
+        hass.bus.async_listen(EVENT_STATE_CHANGED, handle_state_change)
+
+        # Register services
+        async def async_ask_question(call: ServiceCall) -> None:
+            """Handle the ask_question service call."""
+            question = call.data.get("question", "")
+            if not question:
+                _LOGGER.error("No question provided in service call")
+                return
+
+            request_params = {}
+            for param in ["system_prompt", "model", "temperature", "max_tokens"]:
+                if param in call.data:
+                    request_params[param] = call.data[param]
+
+            try:
+                await coordinator.async_ask_question(question, **request_params)
+            except Exception as err:
+                _LOGGER.error("Error asking question: %s", str(err))
+
+        async def async_clear_history(call: ServiceCall) -> None:
+            """Handle the clear_history service call."""
+            try:
+                coordinator._responses.clear()
+                await coordinator.async_refresh()
+                _LOGGER.info("History cleared successfully")
+            except Exception as err:
+                _LOGGER.error("Error clearing history: %s", str(err))
+
+        async def async_get_history(call: ServiceCall) -> dict:
+            """Handle the get_history service call."""
+            try:
+                limit = min(int(call.data.get("limit", 10)), 100)
+                filter_model = str(call.data.get("filter_model", ""))
+                start_date = call.data.get("start_date")
+                include_metadata = call.data.get("include_metadata", False)
+
+                responses = coordinator._responses
+                metrics = {
+                    "total_requests": coordinator.request_count,
+                    "total_tokens": coordinator.tokens_used,
+                    "api_version": coordinator.api_version,
+                    "endpoint_status": coordinator.endpoint_status,
+                    "error_count": coordinator.error_count
+                }
+
+                filtered_responses = responses.copy()
+
+                if filter_model:
+                    filtered_responses = {
+                        k: v for k, v in filtered_responses.items()
+                        if v.get("model") == filter_model
+                    }
+
+                if start_date:
+                    filtered_responses = {
+                        k: v for k, v in filtered_responses.items()
+                        if v.get("timestamp") >= start_date
+                    }
+
+                if not include_metadata:
+                    filtered_responses = {
+                        k: {
+                            "question": v["question"],
+                            "response": v["response"],
+                            "timestamp": v["timestamp"]
+                        } for k, v in filtered_responses.items()
+                    }
+
+                sorted_responses = dict(
+                    sorted(
+                        filtered_responses.items(),
+                        key=lambda x: x[1]["timestamp"],
+                        reverse=True
+                    )[:limit]
+                )
+
+                return {
+                    "metrics": metrics,
+                    "responses": sorted_responses
+                }
+            except Exception as err:
+                _LOGGER.error("Error getting history: %s", str(err))
+                return {}
+
+        async def async_set_system_prompt(call: ServiceCall) -> None:
+            """Handle the set_system_prompt service call."""
+            try:
+                prompt = str(call.data.get("prompt", "")).strip()
+                if prompt:
+                    coordinator.system_prompt = prompt
+                    _LOGGER.info("System prompt updated successfully")
+                else:
+                    _LOGGER.error("Empty prompt provided")
+            except Exception as err:
+                _LOGGER.error("Error setting system prompt: %s", str(err))
+
+        # Register services with validation
+        hass.services.async_register(
+            DOMAIN,
+            "ask_question",
+            async_ask_question,
+            schema=SERVICE_SCHEMA_ASK_QUESTION
+        )
+
+        hass.services.async_register(
+            DOMAIN,
+            "clear_history",
+            async_clear_history
+        )
+
+        hass.services.async_register(
+            DOMAIN,
+            "get_history",
+            async_get_history,
+            schema=SERVICE_SCHEMA_GET_HISTORY
+        )
+
+        hass.services.async_register(
+            DOMAIN,
+            "set_system_prompt",
+            async_set_system_prompt,
+            schema=SERVICE_SCHEMA_SET_SYSTEM_PROMPT
+        )
 
         _LOGGER.info(
             "Successfully set up HA Text AI with model: %s",
@@ -75,19 +324,33 @@ async def async_setup_entry(hass: HomeAssistant, entry: ConfigEntry) -> bool:
         return True
 
     except Exception as ex:
-        _LOGGER.exception("Unexpected error setting up entry: %s", str(ex))
-        return False
+        _LOGGER.exception("Setup error: %s", str(ex))
+        raise ConfigEntryNotReady from ex
 
 async def async_unload_entry(hass: HomeAssistant, entry: ConfigEntry) -> bool:
     """Unload a config entry."""
     try:
-        if entry.entry_id not in hass.data.get(DOMAIN, {}):
-            return True
+        coordinator = hass.data[DOMAIN].get(entry.entry_id)
+        if coordinator:
+            # Clear queue and history
+            coordinator._responses.clear()
+            while not coordinator._question_queue.empty():
+                try:
+                    coordinator._question_queue.get_nowait()
+                    coordinator._question_queue.task_done()
+                except Exception:
+                    pass
+
+            # Close connection
+            await coordinator.async_shutdown()
+
+        # Remove services
+        for service in ["ask_question", "clear_history", "get_history", "set_system_prompt"]:
+            hass.services.async_remove(DOMAIN, service)
 
         unload_ok = await hass.config_entries.async_unload_platforms(entry, PLATFORMS)
         if unload_ok:
-            coordinator = hass.data[DOMAIN].pop(entry.entry_id)
-            await coordinator.async_shutdown()
+            hass.data[DOMAIN].pop(entry.entry_id)
 
         return unload_ok
 
@@ -95,10 +358,18 @@ async def async_unload_entry(hass: HomeAssistant, entry: ConfigEntry) -> bool:
         _LOGGER.exception("Error unloading entry: %s", str(ex))
         return False
 
-async def async_reload_entry(hass: HomeAssistant, entry: ConfigEntry) -> None:
-    """Reload config entry."""
-    try:
-        await async_unload_entry(hass, entry)
-        await async_setup_entry(hass, entry)
-    except Exception as ex:
-        _LOGGER.exception("Error reloading entry: %s", str(ex))
+async def async_migrate_entry(hass: HomeAssistant, entry: ConfigEntry) -> bool:
+    """Migrate old entry."""
+    _LOGGER.debug("Migrating from version %s", entry.version)
+
+    if entry.version == 1:
+        new = {**entry.data}
+
+        # Migrate settings
+        if CONF_MODEL in new and new[CONF_MODEL] not in SUPPORTED_MODELS:
+            new[CONF_MODEL] = DEFAULT_MODEL
+
+        entry.version = 2
+        hass.config_entries.async_update_entry(entry, data=new)
+
+    return True

custom_components/ha_text_ai/config_flow.py

@@ -1,19 +1,17 @@
 """Config flow for HA text AI integration."""
 from typing import Any, Dict, Optional, Tuple
 import voluptuous as vol
-import ssl
-import certifi
 import asyncio
-from async_timeout import timeout
 import aiohttp
-from urllib.parse import urlparse
+from async_timeout import timeout
+from urllib.parse import urlparse, urljoin
 
 from homeassistant import config_entries
 from homeassistant.const import CONF_API_KEY
 import homeassistant.helpers.config_validation as cv
 from homeassistant.core import callback
-from openai import AsyncOpenAI
-from openai import OpenAIError, APIError, APIConnectionError, AuthenticationError, RateLimitError
+from homeassistant.data_entry_flow import FlowResult
+from homeassistant.helpers.aiohttp_client import async_get_clientsession
 
 from .const import (
     DOMAIN,
@@ -27,42 +25,59 @@ from .const import (
     DEFAULT_MAX_TOKENS,
     DEFAULT_API_ENDPOINT,
     DEFAULT_REQUEST_INTERVAL,
+    MIN_TEMPERATURE,
+    MAX_TEMPERATURE,
+    MIN_MAX_TOKENS,
+    MAX_MAX_TOKENS,
+    MIN_REQUEST_INTERVAL,
+    API_VERSION,
+    API_MODELS_PATH,
+    ERROR_INVALID_API_KEY,
+    ERROR_CANNOT_CONNECT,
+    ERROR_UNKNOWN,
+    ERROR_INVALID_MODEL,
+    ERROR_RATE_LIMIT,
+    ERROR_API_ERROR,
+    ERROR_TIMEOUT,
 )
 
 import logging
 _LOGGER = logging.getLogger(__name__)
 
-# Create SSL context at module level
-SSL_CONTEXT = ssl.create_default_context(cafile=certifi.where())
-
-STEP_USER_DATA_SCHEMA = vol.Schema({
-    vol.Required(CONF_API_KEY): str,
-    vol.Optional(CONF_MODEL, default=DEFAULT_MODEL): str,
-    vol.Optional(
-        CONF_TEMPERATURE,
-        default=DEFAULT_TEMPERATURE
-    ): vol.All(
-        vol.Coerce(float),
-        vol.Range(min=0, max=2)
-    ),
-    vol.Optional(
-        CONF_MAX_TOKENS,
-        default=DEFAULT_MAX_TOKENS
-    ): vol.All(
-        vol.Coerce(int),
-        vol.Range(min=1, max=4096)
-    ),
-    vol.Optional(CONF_API_ENDPOINT, default=DEFAULT_API_ENDPOINT): str,
-    vol.Optional(
-        CONF_REQUEST_INTERVAL,
-        default=DEFAULT_REQUEST_INTERVAL
-    ): vol.All(
-        vol.Coerce(float),
-        vol.Range(min=0.1)
-    ),
-})
+STEP_USER_DATA_SCHEMA = vol.Schema(
+    {
+        vol.Required(CONF_API_KEY): cv.string,
+        vol.Optional(CONF_MODEL, default=DEFAULT_MODEL): cv.string,
+        vol.Optional(
+            CONF_TEMPERATURE,
+            default=DEFAULT_TEMPERATURE
+        ): vol.All(
+            vol.Coerce(float),
+            vol.Range(min=MIN_TEMPERATURE, max=MAX_TEMPERATURE)
+        ),
+        vol.Optional(
+            CONF_MAX_TOKENS,
+            default=DEFAULT_MAX_TOKENS
+        ): vol.All(
+            vol.Coerce(int),
+            vol.Range(min=MIN_MAX_TOKENS, max=MAX_MAX_TOKENS)
+        ),
+        vol.Optional(
+            CONF_API_ENDPOINT,
+            default=DEFAULT_API_ENDPOINT
+        ): cv.string,
+        vol.Optional(
+            CONF_REQUEST_INTERVAL,
+            default=DEFAULT_REQUEST_INTERVAL
+        ): vol.All(
+            vol.Coerce(float),
+            vol.Range(min=MIN_REQUEST_INTERVAL)
+        ),
+    }
+)
 
 async def validate_api_connection(
+    hass,
     api_key: str,
     endpoint: str,
     model: str,
@@ -70,25 +85,56 @@ async def validate_api_connection(
     retry_delay: float = 1.0
 ) -> Tuple[bool, str, list]:
     """Validate API connection with retry logic."""
+    session = async_get_clientsession(hass)
+    headers = {
+        "Authorization": f"Bearer {api_key}",
+        "Content-Type": "application/json; charset=utf-8",
+        "Accept": "application/json; charset=utf-8",
+        "Accept-Charset": "utf-8"
+    }
+
+    base_url = endpoint.rstrip('/')
+    if not base_url.endswith(f"/{API_VERSION}"):
+        base_url = f"{base_url}/{API_VERSION}"
+
+    models_url = f"{base_url}/{API_MODELS_PATH}"
+    _LOGGER.debug("Attempting to connect to: %s", models_url)
+
     for attempt in range(retry_count):
         try:
             async with timeout(10):
-                client = AsyncOpenAI(
-                    api_key=api_key,
-                    base_url=endpoint,
-                )
+                async with session.get(models_url, headers=headers) as response:
+                    if response.status == 200:
+                        data = await response.json()
+                        model_ids = [m["id"] for m in data.get("data", [])]
 
-                models = await client.models.list()
-                model_ids = [model.id for model in models.data]
+                        _LOGGER.debug("Available models: %s", ", ".join(model_ids))
 
-                if model not in model_ids:
-                    _LOGGER.warning(
-                        "Model %s not found in available models: %s",
-                        model,
-                        ", ".join(model_ids)
-                    )
-                    return False, "invalid_model", model_ids
-                return True, "", model_ids
+                        if model not in model_ids:
+                            _LOGGER.warning(
+                                "Model %s not found in available models: %s",
+                                model,
+                                ", ".join(model_ids)
+                            )
+                            return False, ERROR_INVALID_MODEL, model_ids
+                        return True, "", model_ids
+
+                    elif response.status == 401:
+                        _LOGGER.error("Authentication failed")
+                        return False, ERROR_INVALID_API_KEY, []
+
+                    elif response.status == 429:
+                        _LOGGER.error("Rate limit exceeded")
+                        return False, ERROR_RATE_LIMIT, []
+
+                    else:
+                        response_text = await response.text()
+                        _LOGGER.error(
+                            "API error: %s - %s",
+                            response.status,
+                            response_text
+                        )
+                        return False, ERROR_API_ERROR, []
 
         except asyncio.TimeoutError:
             _LOGGER.warning(
@@ -97,28 +143,18 @@ async def validate_api_connection(
                 retry_count
             )
             if attempt == retry_count - 1:
-                return False, "timeout", []
+                return False, ERROR_TIMEOUT, []
             await asyncio.sleep(retry_delay)
 
-        except AuthenticationError as err:
-            _LOGGER.error("Authentication error: %s", str(err))
-            return False, "invalid_auth", []
-
-        except RateLimitError as err:
-            _LOGGER.error("Rate limit exceeded: %s", str(err))
-            return False, "rate_limit", []
-
-        except APIConnectionError as err:
-            _LOGGER.error("API connection error: %s", str(err))
-            return False, "cannot_connect", []
-
-        except APIError as err:
-            _LOGGER.error("API error: %s", str(err))
-            return False, "api_error", []
+        except aiohttp.ClientError as err:
+            _LOGGER.error("Connection error: %s", str(err))
+            return False, ERROR_CANNOT_CONNECT, []
 
         except Exception as err:
             _LOGGER.exception("Unexpected error during validation: %s", str(err))
-            return False, "unknown", []
+            return False, ERROR_UNKNOWN, []
+
+    return False, ERROR_UNKNOWN, []
 
 class HATextAIConfigFlow(config_entries.ConfigFlow, domain=DOMAIN):
     """Handle a config flow for HA text AI."""
@@ -128,13 +164,12 @@ class HATextAIConfigFlow(config_entries.ConfigFlow, domain=DOMAIN):
     async def async_step_user(
         self,
         user_input: Optional[Dict[str, Any]] = None
-    ) -> Dict[str, Any]:
+    ) -> FlowResult:
         """Handle the initial step."""
         errors: Dict[str, str] = {}
 
         if user_input is not None:
             try:
-                # Validate URL format
                 endpoint = user_input.get(CONF_API_ENDPOINT, DEFAULT_API_ENDPOINT)
                 try:
                     result = urlparse(endpoint)
@@ -154,29 +189,29 @@ class HATextAIConfigFlow(config_entries.ConfigFlow, domain=DOMAIN):
                         errors=errors
                     )
 
-                # Validate input data
-                user_input = STEP_USER_DATA_SCHEMA(user_input)
+                validated_input = STEP_USER_DATA_SCHEMA(user_input)
 
                 is_valid, error_code, available_models = await validate_api_connection(
-                    user_input[CONF_API_KEY],
+                    self.hass,
+                    validated_input[CONF_API_KEY],
                     endpoint,
-                    user_input[CONF_MODEL]
+                    validated_input[CONF_MODEL]
                 )
 
                 if is_valid:
-                    await self.async_set_unique_id(user_input[CONF_API_KEY])
+                    await self.async_set_unique_id(validated_input[CONF_API_KEY])
                     self._abort_if_unique_id_configured()
 
                     return self.async_create_entry(
-                        title="HA text AI",
-                        data=user_input
+                        title="HA Text AI",
+                        data=validated_input
                     )
 
                 errors["base"] = error_code
-                if error_code == "invalid_model":
+                if error_code == ERROR_INVALID_MODEL:
                     _LOGGER.warning(
                         "Selected model %s not found in available models: %s",
-                        user_input[CONF_MODEL],
+                        validated_input[CONF_MODEL],
                         ", ".join(available_models)
                     )
 
@@ -212,7 +247,7 @@ class OptionsFlowHandler(config_entries.OptionsFlow):
     async def async_step_init(
         self,
         user_input: Optional[Dict[str, Any]] = None
-    ) -> Dict[str, Any]:
+    ) -> FlowResult:
         """Handle options flow."""
         if user_input is not None:
             return self.async_create_entry(title="", data=user_input)
@@ -223,30 +258,27 @@ class OptionsFlowHandler(config_entries.OptionsFlow):
                 default=self.config_entry.options.get(
                     CONF_TEMPERATURE, DEFAULT_TEMPERATURE
                 ),
-                description={"suggested_value": DEFAULT_TEMPERATURE},
             ): vol.All(
                 vol.Coerce(float),
-                vol.Range(min=0, max=2)
+                vol.Range(min=MIN_TEMPERATURE, max=MAX_TEMPERATURE)
             ),
             vol.Optional(
                 CONF_MAX_TOKENS,
                 default=self.config_entry.options.get(
                     CONF_MAX_TOKENS, DEFAULT_MAX_TOKENS
                 ),
-                description={"suggested_value": DEFAULT_MAX_TOKENS},
             ): vol.All(
                 vol.Coerce(int),
-                vol.Range(min=1, max=4096)
+                vol.Range(min=MIN_MAX_TOKENS, max=MAX_MAX_TOKENS)
             ),
             vol.Optional(
                 CONF_REQUEST_INTERVAL,
                 default=self.config_entry.options.get(
                     CONF_REQUEST_INTERVAL, DEFAULT_REQUEST_INTERVAL
                 ),
-                description={"suggested_value": DEFAULT_REQUEST_INTERVAL},
             ): vol.All(
                 vol.Coerce(float),
-                vol.Range(min=0.1)
+                vol.Range(min=MIN_REQUEST_INTERVAL)
             ),
         })
 

custom_components/ha_text_ai/const.py

@@ -13,15 +13,27 @@ CONF_MAX_TOKENS: Final = "max_tokens"
 CONF_API_ENDPOINT: Final = "api_endpoint"
 CONF_REQUEST_INTERVAL: Final = "request_interval"
 
+# Model constants
+SUPPORTED_MODELS: Final = [
+    "gpt-3.5-turbo",
+    "gpt-3.5-turbo-16k",
+    "gpt-4",
+    "gpt-4-32k",
+    "gpt-4-1106-preview",
+    "claude-3-sonnet",
+    "claude-3-opus"
+]
+
 # Default values
 DEFAULT_MODEL: Final = "gpt-3.5-turbo"
-DEFAULT_TEMPERATURE: Final = 0.7
+DEFAULT_TEMPERATURE: Final = 0.1
 DEFAULT_MAX_TOKENS: Final = 1000
-DEFAULT_API_ENDPOINT: Final = "https://api.openai.com/v1"
+DEFAULT_API_ENDPOINT: Final = "https://api.openai.com"
 DEFAULT_REQUEST_INTERVAL: Final = 1.0
 DEFAULT_TIMEOUT: Final = 30
 DEFAULT_QUEUE_SIZE: Final = 100
 DEFAULT_HISTORY_LIMIT: Final = 50
+DEFAULT_RETRY_COUNT: Final = 3
 
 # Parameter constraints
 MIN_TEMPERATURE: Final = 0.0
@@ -31,6 +43,23 @@ MAX_MAX_TOKENS: Final = 4096
 MIN_REQUEST_INTERVAL: Final = 0.1
 MIN_TIMEOUT: Final = 5
 MAX_TIMEOUT: Final = 120
+MAX_PROMPT_LENGTH: Final = 1000
+MAX_HISTORY_LIMIT: Final = 100
+
+# API constants
+API_VERSION: Final = "v1"
+API_MODELS_PATH: Final = "models"
+API_CHAT_PATH: Final = "chat/completions"
+API_TIMEOUT: Final = 30
+API_RETRY_COUNT: Final = 3
+API_BACKOFF_FACTOR: Final = 1.5
+API_MAX_RETRIES: Final = 3
+
+# History constants
+HISTORY_FILTER_MODEL: Final = "filter_model"
+HISTORY_FILTER_DATE: Final = "start_date"
+HISTORY_SORT_ORDER: Final = "sort_order"
+HISTORY_INCLUDE_METADATA: Final = "include_metadata"
 
 # Service names
 SERVICE_ASK_QUESTION: Final = "ask_question"
@@ -58,6 +87,13 @@ ATTR_QUEUE_SIZE: Final = "queue_size"
 ATTR_API_STATUS: Final = "api_status"
 ATTR_ERROR_COUNT: Final = "error_count"
 ATTR_LAST_ERROR: Final = "last_error"
+ATTR_API_VERSION: Final = "api_version"
+ATTR_ENDPOINT_STATUS: Final = "endpoint_status"
+ATTR_REQUEST_COUNT: Final = "request_count"
+ATTR_TOKENS_USED: Final = "tokens_used"
+ATTR_RETRY_COUNT: Final = "retry_count"
+ATTR_QUEUE_POSITION: Final = "queue_position"
+ATTR_ESTIMATED_WAIT: Final = "estimated_wait"
 
 # Error messages
 ERROR_INVALID_API_KEY: Final = "invalid_api_key"
@@ -70,6 +106,8 @@ ERROR_API_ERROR: Final = "api_error"
 ERROR_TIMEOUT: Final = "timeout_error"
 ERROR_QUEUE_FULL: Final = "queue_full"
 ERROR_INVALID_PROMPT: Final = "invalid_prompt"
+ERROR_INVALID_PARAMETERS: Final = "invalid_parameters"
+ERROR_SERVICE_UNAVAILABLE: Final = "service_unavailable"
 
 # Configuration descriptions
 CONF_MODEL_DESCRIPTION: Final = "AI model to use for responses"
@@ -92,12 +130,18 @@ ATTR_QUEUE_SIZE_DESCRIPTION: Final = "Current size of question queue"
 ATTR_API_STATUS_DESCRIPTION: Final = "Current API connection status"
 ATTR_ERROR_COUNT_DESCRIPTION: Final = "Total number of errors"
 ATTR_LAST_ERROR_DESCRIPTION: Final = "Last error message"
+ATTR_API_VERSION_DESCRIPTION: Final = "Current API version"
+ATTR_ENDPOINT_STATUS_DESCRIPTION: Final = "Current endpoint status"
+ATTR_REQUEST_COUNT_DESCRIPTION: Final = "Total number of API requests"
+ATTR_TOKENS_USED_DESCRIPTION: Final = "Total tokens used"
 
 # Entity attributes
 ENTITY_NAME: Final = "HA Text AI"
 ENTITY_ICON: Final = "mdi:robot"
 ENTITY_ICON_ERROR: Final = "mdi:robot-dead"
 ENTITY_ICON_PROCESSING: Final = "mdi:robot-excited"
+ENTITY_ICON_OFFLINE: Final = "mdi:robot-off"
+ENTITY_ICON_QUEUE: Final = "mdi:robot-confused"
 
 # Translation keys
 TRANSLATION_KEY_CONFIG: Final = "config"
@@ -113,6 +157,10 @@ STATE_ERROR: Final = "error"
 STATE_DISCONNECTED: Final = "disconnected"
 STATE_RATE_LIMITED: Final = "rate_limited"
 STATE_INITIALIZING: Final = "initializing"
+STATE_MAINTENANCE: Final = "maintenance"
+STATE_RETRYING: Final = "retrying"
+STATE_QUEUED: Final = "queued"
+STATE_UPDATING: Final = "updating"
 
 # Logging
 LOGGER_NAME: Final = "custom_components.ha_text_ai"
@@ -122,11 +170,6 @@ LOG_LEVEL_DEFAULT: Final = "INFO"
 QUEUE_TIMEOUT: Final = 5
 QUEUE_MAX_SIZE: Final = 100
 
-# API constants
-API_TIMEOUT: Final = 30
-API_RETRY_COUNT: Final = 3
-API_BACKOFF_FACTOR: Final = 1.5
-
 # Service schema constants
 SCHEMA_QUESTION: Final = "question"
 SCHEMA_MODEL: Final = "model"

custom_components/ha_text_ai/coordinator.py

@@ -2,14 +2,24 @@
 import asyncio
 import logging
 from datetime import timedelta
-from typing import Any, Dict, Optional
+from typing import Any, Dict, Optional, List
+import time
 
 from openai import AsyncOpenAI, APIError, AuthenticationError, RateLimitError
+from anthropic import AsyncAnthropic
 from homeassistant.core import HomeAssistant
 from homeassistant.helpers.update_coordinator import DataUpdateCoordinator
+from homeassistant.util import dt as dt_util
 import async_timeout
 
-from .const import DOMAIN
+from .const import (
+    DOMAIN,
+    DEFAULT_TIMEOUT,
+    MAX_QUEUE_SIZE,
+    MAX_RETRIES,
+    RETRY_DELAY,
+    SUPPORTED_MODELS,
+)
 
 _LOGGER = logging.getLogger(__name__)
 
@@ -26,8 +36,21 @@ class HATextAICoordinator(DataUpdateCoordinator):
         max_tokens: int,
         request_interval: float,
         session: Optional[Any] = None,
+        is_anthropic: bool = False,
     ) -> None:
-        """Initialize."""
+        """Initialize coordinator.
+
+        Args:
+            hass: HomeAssistant instance
+            api_key: API key for the service
+            endpoint: API endpoint URL
+            model: Model name to use
+            temperature: Temperature parameter for generation
+            max_tokens: Maximum tokens to generate
+            request_interval: Interval between requests
+            session: Optional session object
+            is_anthropic: Whether to use Anthropic API
+        """
         super().__init__(
             hass,
             _LOGGER,
@@ -42,18 +65,36 @@ class HATextAICoordinator(DataUpdateCoordinator):
         self.model = model
         self.temperature = float(temperature)
         self.max_tokens = int(max_tokens)
-        self._question_queue = asyncio.Queue()
+        self._question_queue = asyncio.PriorityQueue(maxsize=MAX_QUEUE_SIZE)
         self._responses: Dict[str, Any] = {}
         self.system_prompt: Optional[str] = None
         self._is_ready = False
+        self._is_processing = False
+        self._is_rate_limited = False
+        self._is_maintenance = False
         self._error_count = 0
         self._MAX_ERRORS = 3
+        self._request_count = 0
+        self._tokens_used = 0
+        self._api_version = "v1"
+        self._endpoint_status = "disconnected"
+        self._performance_metrics: Dict[str, Any] = {
+            "avg_response_time": 0,
+            "total_errors": 0,
+            "success_rate": 100,
+            "requests_per_minute": 0,
+        }
+        self._last_request_time = 0
+        self._is_anthropic = is_anthropic
 
-        self.client = AsyncOpenAI(
-            api_key=self.api_key,
-            base_url=self.endpoint,
-            http_client=session,
-        )
+        if is_anthropic:
+            self.client = AsyncAnthropic(api_key=self.api_key)
+        else:
+            self.client = AsyncOpenAI(
+                api_key=self.api_key,
+                base_url=self.endpoint,
+                http_client=session,
+            )
 
     def _validate_params(self, api_key: str, temperature: float, max_tokens: int) -> None:
         """Validate initialization parameters."""
@@ -65,28 +106,59 @@ class HATextAICoordinator(DataUpdateCoordinator):
             raise ValueError("Max tokens must be a positive integer")
 
     async def _async_update_data(self) -> Dict[str, Any]:
-        """Update data via OpenAI API."""
+        """Update data via API."""
         if self._question_queue.empty():
             return self._responses
 
         try:
-            async with async_timeout.timeout(30):
-                question = await self._question_queue.get()
+            async with async_timeout.timeout(DEFAULT_TIMEOUT):
+                self._is_processing = True
+                priority, question_data = await self._question_queue.get()
+                question = question_data["question"]
+                params = question_data["params"]
+
                 try:
-                    response_content = await self._make_api_call(question)
+                    response_data = await self._make_api_call(
+                        question,
+                        model=params.get("model"),
+                        temperature=params.get("temperature"),
+                        max_tokens=params.get("max_tokens"),
+                        system_prompt=params.get("system_prompt")
+                    )
+
+                    self._update_metrics(response_data)
                     self._responses[question] = {
                         "question": question,
-                        "response": response_content,
+                        "response": response_data["response"],
                         "error": None,
-                        "timestamp": self.hass.loop.time()
+                        "timestamp": dt_util.utcnow(),
+                        "model": response_data["model"],
+                        "temperature": params.get("temperature", self.temperature),
+                        "max_tokens": params.get("max_tokens", self.max_tokens),
+                        "response_time": response_data.get("response_time"),
+                        "tokens": response_data.get("tokens", 0),
+                        "priority": priority
                     }
                     self._error_count = 0
                     self._is_ready = True
+                    self._endpoint_status = "connected"
+                    self._request_count += 1
+                    self._tokens_used += response_data.get("tokens", 0)
+                    self._last_request_time = time.time()
+
+                    # Fire event for successful response
+                    self.hass.bus.async_fire(f"{DOMAIN}_response_received", {
+                        "question": question,
+                        "model": response_data["model"],
+                        "tokens": response_data.get("tokens", 0)
+                    })
+
                     _LOGGER.debug("Response received for question: %s", question)
 
                 except Exception as err:
-                    self._handle_api_error(question, err)
+                    await self._handle_api_error(question, err)
                 finally:
+                    self._is_processing = False
                     self._question_queue.task_done()
 
                 return self._responses
@@ -96,26 +168,47 @@ class HATextAICoordinator(DataUpdateCoordinator):
             await self._handle_timeout_error()
             return self._responses
 
-    def _handle_api_error(self, question: str, error: Exception) -> None:
-        """Handle API errors."""
+    async def _handle_api_error(self, question: str, error: Exception) -> None:
+        """Handle API errors with retry logic."""
         self._error_count += 1
+        self._performance_metrics["total_errors"] += 1
         error_msg = str(error)
 
         if isinstance(error, AuthenticationError):
             error_msg = "Authentication failed - invalid API key"
             self._is_ready = False
+            self._endpoint_status = "auth_error"
         elif isinstance(error, RateLimitError):
             error_msg = "Rate limit exceeded"
+            self._is_rate_limited = True
+            self._endpoint_status = "rate_limited"
+            # Implement exponential backoff
+            await asyncio.sleep(RETRY_DELAY * (2 ** (self._error_count - 1)))
         elif isinstance(error, APIError):
+            if "maintenance" in str(error).lower():
+                self._is_maintenance = True
+                self._endpoint_status = "maintenance"
             error_msg = f"API error: {error}"
+        else:
+            self._endpoint_status = "error"
 
         self._responses[question] = {
             "question": question,
             "response": None,
             "error": error_msg,
-            "timestamp": self.hass.loop.time()
+            "timestamp": dt_util.utcnow(),
+            "model": self.model,
+            "temperature": self.temperature,
+            "max_tokens": self.max_tokens
         }
 
+        # Fire error event
+        self.hass.bus.async_fire(f"{DOMAIN}_error_occurred", {
+            "error_type": type(error).__name__,
+            "error_message": error_msg,
+            "question": question
+        })
+
         _LOGGER.error("API error (%s): %s", type(error).__name__, error_msg)
 
         if self._error_count >= self._MAX_ERRORS:
@@ -124,71 +217,402 @@ class HATextAICoordinator(DataUpdateCoordinator):
                 self._error_count
             )
 
-    async def _handle_timeout_error(self) -> None:
-        """Handle timeout errors."""
-        self._error_count += 1
-        if not self._question_queue.empty():
-            try:
-                # Clear the queue if we have timeout issues
-                while not self._question_queue.empty():
-                    self._question_queue.get_nowait()
-                    self._question_queue.task_done()
-            except Exception as err:
-                _LOGGER.error("Error clearing question queue: %s", err)
+    def _update_metrics(self, response_data: Dict[str, Any]) -> None:
+        """Update performance metrics."""
+        response_time = response_data.get("response_time", 0)
+        current_avg = self._performance_metrics["avg_response_time"]
+        self._performance_metrics["avg_response_time"] = (
+            (current_avg * self._request_count + response_time) /
+            (self._request_count + 1)
+        )
 
-    async def _make_api_call(self, question: str) -> str:
-        """Make API call to OpenAI."""
+        total_requests = self._request_count + 1
+        self._performance_metrics["success_rate"] = (
+            (total_requests - self._performance_metrics["total_errors"]) /
+            total_requests * 100
+        )
+
+        # Calculate requests per minute
+        if self._last_request_time:
+            time_diff = time.time() - self._last_request_time
+            if time_diff > 0:
+                self._performance_metrics["requests_per_minute"] = 60 / time_diff
+
+    async def _make_api_call(
+        self,
+        question: str,
+        model: Optional[str] = None,
+        temperature: Optional[float] = None,
+        max_tokens: Optional[int] = None,
+        system_prompt: Optional[str] = None
+    ) -> Dict[str, Any]:
+        """Make API call to the selected service."""
         try:
-            messages = []
-            if self.system_prompt:
-                messages.append({"role": "system", "content": self.system_prompt})
-            messages.append({"role": "user", "content": question})
+            start_time = dt_util.utcnow()
 
-            completion = await self.client.chat.completions.create(
-                model=self.model,
-                messages=messages,
-                temperature=self.temperature,
-                max_tokens=self.max_tokens,
-            )
-            return completion.choices[0].message.content
+            if self._is_anthropic:
+                response = await self._make_anthropic_call(
+                    question, model, temperature, max_tokens, system_prompt
+                )
+            else:
+                response = await self._make_openai_call(
+                    question, model, temperature, max_tokens, system_prompt
+                )
+
+            response_time = (dt_util.utcnow() - start_time).total_seconds()
+
+            return {
+                **response,
+                "response_time": response_time
+            }
 
         except Exception as err:
             _LOGGER.error("Error in API call: %s", err)
             raise
 
-    async def async_ask_question(self, question: str) -> None:
-        """Add question to queue."""
+    async def _make_anthropic_call(
+        self,
+        question: str,
+        model: Optional[str],
+        temperature: Optional[float],
+        max_tokens: Optional[int],
+        system_prompt: Optional[str]
+    ) -> Dict[str, Any]:
+        """Make API call to Anthropic."""
+        messages = []
+        if system_prompt:
+            messages.append({"role": "system", "content": system_prompt})
+        messages.append({"role": "user", "content": question})
+
+        completion = await self.client.messages.create(
+            model=model or self.model,
+            messages=messages,
+            temperature=temperature if temperature is not None else self.temperature,
+            max_tokens=max_tokens if max_tokens is not None else self.max_tokens,
+        )
+
+        return {
+            "response": completion.content[0].text,
+            "model": completion.model,
+            "tokens": completion.usage.total_tokens if hasattr(completion, 'usage') else 0
+        }
+
+    async def _make_openai_call(
+        self,
+        question: str,
+        model: Optional[str],
+        temperature: Optional[float],
+        max_tokens: Optional[int],
+        system_prompt: Optional[str]
+    ) -> Dict[str, Any]:
+        """Make API call to OpenAI."""
+        messages = []
+        if system_prompt:
+            messages.append({"role": "system", "content": system_prompt})
+        messages.append({"role": "user", "content": question})
+
+        completion = await self.client.chat.completions.create(
+            model=model or self.model,
+            messages=messages,
+            temperature=temperature if temperature is not None else self.temperature,
+            max_tokens=max_tokens if max_tokens is not None else self.max_tokens,
+        )
+
+        return {
+            "response": completion.choices[0].message.content,
+            "model": completion.model,
+            "tokens": completion.usage.total_tokens if hasattr(completion, 'usage') else 0
+        }
+
+    async def async_ask_question(
+        self,
+        question: str,
+        system_prompt: Optional[str] = None,
+        model: Optional[str] = None,
+        temperature: Optional[float] = None,
+        max_tokens: Optional[int] = None,
+        priority: bool = False
+    ) -> None:
+        """Add question to queue with priority support."""
         if not self._is_ready and self._error_count >= self._MAX_ERRORS:
             _LOGGER.warning("Coordinator is not ready due to previous errors")
             return
 
-        await self._question_queue.put(question)
-        await self.async_refresh()
+        question_data = {
+            "question": question,
+            "params": {
+                "system_prompt": system_prompt,
+                "model": model,
+                "temperature": temperature,
+                "max_tokens": max_tokens
+            }
+        }
+
+        # Priority: 0 for high priority, 1 for normal
+        priority_level = 0 if priority else 1
+
+        try:
+            await self._question_queue.put((priority_level, question_data))
+            await self.async_refresh()
+        except asyncio.QueueFull:
+            _LOGGER.error("Question queue is full. Try again later.")
+            raise RuntimeError("Queue is full")
 
     async def async_shutdown(self) -> None:
         """Shutdown the coordinator."""
         try:
-            # Clear the queue
             while not self._question_queue.empty():
-                self._question_queue.get_nowait()
-                self._question_queue.task_done()
+                try:
+                    self._question_queue.get_nowait()
+                    self._question_queue.task_done()
+                except asyncio.QueueEmpty:
+                    break
+
+            if hasattr(self.client, 'close'):
+                await self.client.close()
 
-            await self.client.close()
             self._is_ready = False
+            self._endpoint_status = "disconnected"
+
+            # Final metrics update
+            self._update_final_metrics()
 
         except Exception as err:
             _LOGGER.error("Error during shutdown: %s", err)
 
+    def _update_final_metrics(self) -> None:
+        """Update final metrics before shutdown."""
+        if self._request_count > 0:
+            self._performance_metrics["final_success_rate"] = (
+                (self._request_count - self._performance_metrics["total_errors"]) /
+                self._request_count * 100
+            )
+            self._performance_metrics["total_requests"] = self._request_count
+            self._performance_metrics["total_tokens"] = self._tokens_used
+
+    @property
+    def performance_metrics(self) -> Dict[str, Any]:
+        """Return current performance metrics."""
+        return self._performance_metrics
+
+    @property
+    def queue_size(self) -> int:
+        """Return current queue size."""
+        return self._question_queue.qsize()
+
+    @property
+    def is_queue_full(self) -> bool:
+        """Return whether queue is full."""
+        return self._question_queue.full()
+
     @property
     def is_ready(self) -> bool:
         """Return if coordinator is ready."""
-        return self._is_ready
+        return self._is_ready and self._error_count < self._MAX_ERRORS
+
+    @property
+    def is_processing(self) -> bool:
+        """Return if coordinator is processing."""
+        return self._is_processing
+
+    @property
+    def is_rate_limited(self) -> bool:
+        """Return if coordinator is rate limited."""
+        return self._is_rate_limited
+
+    @property
+    def is_maintenance(self) -> bool:
+        """Return if API is in maintenance."""
+        return self._is_maintenance
 
     @property
     def error_count(self) -> int:
         """Return current error count."""
         return self._error_count
 
+    @property
+    def request_count(self) -> int:
+        """Return total request count."""
+        return self._request_count
+
+    @property
+    def tokens_used(self) -> int:
+        """Return total tokens used."""
+        return self._tokens_used
+
+    @property
+    def api_version(self) -> str:
+        """Return API version."""
+        return self._api_version
+
+    @property
+    def endpoint_status(self) -> str:
+        """Return endpoint status."""
+        return self._endpoint_status
+
+    @property
+    def responses(self) -> Dict[str, Any]:
+        """Return all responses."""
+        return self._responses
+
+    @property
+    def last_response(self) -> Optional[Dict[str, Any]]:
+        """Return the last response."""
+        if not self._responses:
+            return None
+        return next(iter(self._responses.values()))
+
     def reset_error_count(self) -> None:
         """Reset error counter."""
         self._error_count = 0
+        self._is_rate_limited = False
+        self._is_maintenance = False
+        if not self._is_ready:
+            self._is_ready = True
+            self._endpoint_status = "connected"
+
+    async def clear_queue(self) -> None:
+        """Clear the question queue."""
+        try:
+            while not self._question_queue.empty():
+                try:
+                    self._question_queue.get_nowait()
+                    self._question_queue.task_done()
+                except asyncio.QueueEmpty:
+                    break
+        except Exception as err:
+            _LOGGER.error("Error clearing queue: %s", err)
+
+    async def clear_history(self) -> None:
+        """Clear response history."""
+        self._responses.clear()
+        await self.async_refresh()
+
+    def get_response(self, question: str) -> Optional[Dict[str, Any]]:
+        """Get specific response by question."""
+        return self._responses.get(question)
+
+    def get_recent_responses(self, limit: int = 10) -> List[Dict[str, Any]]:
+        """Get most recent responses."""
+        return list(sorted(
+            self._responses.values(),
+            key=lambda x: x["timestamp"],
+            reverse=True
+        ))[:limit]
+
+    async def retry_failed_requests(self) -> None:
+        """Retry failed requests."""
+        failed_requests = [
+            (q, r) for q, r in self._responses.items()
+            if r.get("error") is not None
+        ]
+
+        for question, response in failed_requests:
+            await self.async_ask_question(
+                question,
+                system_prompt=response.get("system_prompt"),
+                model=response.get("model"),
+                temperature=response.get("temperature"),
+                max_tokens=response.get("max_tokens"),
+                priority=True
+            )
+
+    def update_system_prompt(self, new_prompt: str) -> None:
+        """Update system prompt."""
+        self.system_prompt = new_prompt
+        _LOGGER.info("System prompt updated")
+
+    async def health_check(self) -> Dict[str, Any]:
+        """Perform health check."""
+        health_status = {
+            "is_ready": self.is_ready,
+            "is_processing": self.is_processing,
+            "is_rate_limited": self.is_rate_limited,
+            "is_maintenance": self.is_maintenance,
+            "error_count": self.error_count,
+            "endpoint_status": self.endpoint_status,
+            "queue_size": self.queue_size,
+            "request_count": self.request_count,
+            "tokens_used": self.tokens_used,
+            "performance_metrics": self.performance_metrics
+        }
+
+        return health_status
+
+    async def _handle_timeout_error(self) -> None:
+        """Handle timeout errors."""
+        self._error_count += 1
+        self._endpoint_status = "timeout"
+        self._performance_metrics["total_errors"] += 1
+
+        if not self._question_queue.empty():
+            await self.clear_queue()
+
+        # Fire timeout event
+        self.hass.bus.async_fire(f"{DOMAIN}_timeout_error", {
+            "error_count": self._error_count,
+            "endpoint_status": self._endpoint_status
+        })
+
+    def export_metrics(self) -> Dict[str, Any]:
+        """Export all metrics and statistics."""
+        return {
+            "performance": self._performance_metrics,
+            "requests": {
+                "total": self._request_count,
+                "successful": self._request_count - self._performance_metrics["total_errors"],
+                "failed": self._performance_metrics["total_errors"]
+            },
+            "tokens": {
+                "total_used": self._tokens_used,
+                "average_per_request": self._tokens_used / self._request_count if self._request_count > 0 else 0
+            },
+            "status": {
+                "is_ready": self.is_ready,
+                "endpoint_status": self._endpoint_status,
+                "error_count": self._error_count
+            },
+            "queue": {
+                "size": self.queue_size,
+                "is_full": self.is_queue_full
+            }
+        }
+
+    async def validate_model(self, model: str) -> bool:
+        """Validate if model is supported."""
+        return model in SUPPORTED_MODELS
+
+    async def estimate_tokens(self, text: str) -> int:
+        """Estimate token count for text."""
+        # Простая оценка: примерно 4 символа на токен
+        return len(text) // 4
+
+    def get_rate_limit_info(self) -> Dict[str, Any]:
+        """Get rate limit information."""
+        return {
+            "is_rate_limited": self._is_rate_limited,
+            "retry_after": RETRY_DELAY * (2 ** (self._error_count - 1)) if self._is_rate_limited else 0
+        }
+
+    async def optimize_queue(self) -> None:
+        """Optimize queue by removing duplicate requests."""
+        if self._question_queue.empty():
+            return
+
+        seen_questions = set()
+        optimized_queue = asyncio.PriorityQueue(maxsize=MAX_QUEUE_SIZE)
+
+        while not self._question_queue.empty():
+            try:
+                priority, question_data = self._question_queue.get_nowait()
+                question = question_data["question"]
+
+                if question not in seen_questions:
+                    seen_questions.add(question)
+                    await optimized_queue.put((priority, question_data))
+
+                self._question_queue.task_done()
+            except asyncio.QueueEmpty:
+                break
+
+        self._question_queue = optimized_queue

custom_components/ha_text_ai/manifest.json

@@ -4,11 +4,39 @@
   "codeowners": ["@smkrv"],
   "config_flow": true,
   "dependencies": [],
-  "documentation": "https://github.com/smkrv/ha-text-ai/wiki",
+  "documentation": "https://github.com/smkrv/ha-text-ai",
+  "homepage": "https://github.com/smkrv/ha-text-ai",
   "iot_class": "cloud_polling",
   "issue_tracker": "https://github.com/smkrv/ha-text-ai/issues",
-  "requirements": ["openai>=1.0.0"],
+  "requirements": [
+    "openai>=1.12.0",
+    "anthropic>=0.8.0",
+    "aiohttp>=3.8.0",
+    "async-timeout>=4.0.0"
+  ],
+  "version": "2.0.0",
+  "homeassistant": "2024.1.0",
+  "quality_scale": "silver",
+  "integration_type": "service",
+  "translations": [
+    "en"
+  ],
+  "zeroconf": [],
   "ssdp": [],
-  "version": "1.0.9",
-  "zeroconf": []
+  "usb": [],
+  "bluetooth": [],
+  "mqtt": [],
+  "loggers": ["custom_components.ha_text_ai"],
+  "after_dependencies": ["http"],
+  "title": "Text AI Assistant",
+  "description": "AI-powered text assistant supporting multiple models including GPT and Claude",
+  "render_readme": true,
+  "tags": [
+    "ai",
+    "automation",
+    "chat",
+    "assistant",
+    "openai",
+    "claude"
+  ]
 }

custom_components/ha_text_ai/sensor.py

@@ -30,6 +30,10 @@ from .const import (
     ATTR_ERROR_COUNT,
     ATTR_LAST_ERROR,
     ATTR_RESPONSE_TIME,
+    ATTR_API_VERSION,
+    ATTR_ENDPOINT_STATUS,
+    ATTR_REQUEST_COUNT,
+    ATTR_TOKENS_USED,
     ENTITY_ICON,
     ENTITY_ICON_ERROR,
     ENTITY_ICON_PROCESSING,
@@ -39,6 +43,7 @@ from .const import (
     STATE_DISCONNECTED,
     STATE_RATE_LIMITED,
     STATE_INITIALIZING,
+    STATE_MAINTENANCE,
 )
 from .coordinator import HATextAICoordinator
 
@@ -53,6 +58,7 @@ async def async_setup_entry(
     coordinator = hass.data[DOMAIN][entry.entry_id]
     async_add_entities([HATextAISensor(coordinator, entry)], True)
 
+
 class HATextAISensor(CoordinatorEntity, SensorEntity):
     """HA text AI Sensor."""
 
@@ -74,6 +80,13 @@ class HATextAISensor(CoordinatorEntity, SensorEntity):
         self._error_count = 0
         self._last_error = None
         self._state = STATE_INITIALIZING
+        self._attr_device_info = {
+            "identifiers": {(DOMAIN, self._attr_unique_id)},
+            "name": "HA Text AI",
+            "manufacturer": "Community",
+            "model": coordinator.model,
+            "sw_version": coordinator.api_version,
+        }
 
     @property
     def icon(self) -> str:
@@ -87,13 +100,16 @@ class HATextAISensor(CoordinatorEntity, SensorEntity):
     @property
     def state(self) -> StateType:
         """Return the state of the sensor."""
-        if not self.coordinator.data or not self.coordinator.last_update_success_time:
+        if not self.coordinator.data or not self.coordinator.last_update_success:
             return None
 
         try:
-            if isinstance(self.coordinator.last_update_success_time, datetime):
-                return dt_util.as_local(self.coordinator.last_update_success_time)
-            return self.coordinator.last_update_success_time
+            if self.coordinator.data and isinstance(self.coordinator.data, dict):
+                last_update = self.coordinator.data.get("last_update")
+                if isinstance(last_update, datetime):
+                    return dt_util.as_local(last_update)
+                return last_update
+            return None
         except Exception as err:
             _LOGGER.error("Error getting state: %s", err, exc_info=True)
             return None
@@ -111,27 +127,41 @@ class HATextAISensor(CoordinatorEntity, SensorEntity):
             ATTR_API_STATUS: self._state,
             ATTR_ERROR_COUNT: self._error_count,
             ATTR_LAST_ERROR: self._last_error,
+            ATTR_API_VERSION: self.coordinator.api_version,
+            ATTR_ENDPOINT_STATUS: self.coordinator.endpoint_status,
+            ATTR_REQUEST_COUNT: self.coordinator.request_count,
+            ATTR_TOKENS_USED: self.coordinator.tokens_used,
         }
 
         if not self.coordinator.data:
             return attributes
 
         try:
-            history = list(self.coordinator.data.items())
+            history = list(self.coordinator._responses.items())
             if history:
                 last_question, last_data = history[-1]
 
-                # Handle different response formats
                 if isinstance(last_data, dict):
                     last_response = last_data.get("response", "")
-                    last_updated = last_data.get("timestamp", self.coordinator.last_update_success_time)
+                    last_updated = last_data.get("timestamp") or self.coordinator.data.get("last_update")
                     response_time = last_data.get("response_time")
+
+                    model = last_data.get("model", self.coordinator.model)
+                    temperature = last_data.get("temperature", self.coordinator.temperature)
+                    max_tokens = last_data.get("max_tokens", self.coordinator.max_tokens)
+                    error = last_data.get("error")
+
+                    if error:
+                        self._last_error = error
+                        self._state = STATE_ERROR
                 else:
                     last_response = str(last_data)
-                    last_updated = self.coordinator.last_update_success_time
+                    last_updated = self.coordinator.data.get("last_update")
                     response_time = None
+                    model = self.coordinator.model
+                    temperature = self.coordinator.temperature
+                    max_tokens = self.coordinator.max_tokens
 
-                # Convert timestamp to local time if needed
                 if isinstance(last_updated, datetime):
                     last_updated = dt_util.as_local(last_updated)
 
@@ -140,6 +170,9 @@ class HATextAISensor(CoordinatorEntity, SensorEntity):
                     ATTR_RESPONSE: last_response,
                     ATTR_LAST_UPDATED: last_updated,
                     ATTR_TOTAL_RESPONSES: len(history),
+                    ATTR_MODEL: model,
+                    ATTR_TEMPERATURE: temperature,
+                    ATTR_MAX_TOKENS: max_tokens,
                 })
 
                 if response_time is not None:
@@ -169,9 +202,24 @@ class HATextAISensor(CoordinatorEntity, SensorEntity):
         """Handle updated data from the coordinator."""
         try:
             if self.coordinator.data:
-                self._state = STATE_READY
+                if self.coordinator._is_ready:
+                    if self.coordinator._is_processing:
+                        self._state = STATE_PROCESSING
+                    elif self.coordinator._is_rate_limited:
+                        self._state = STATE_RATE_LIMITED
+                    elif self.coordinator._is_maintenance:
+                        self._state = STATE_MAINTENANCE
+                    else:
+                        self._state = STATE_READY
+                else:
+                    self._state = STATE_DISCONNECTED
             else:
                 self._state = STATE_DISCONNECTED
+
+            # Обновляем счетчик ошибок только если статус изменился на ошибку
+            if self._state == STATE_ERROR:
+                self._error_count += 1
+
         except Exception as err:
             _LOGGER.error("Error handling update: %s", err, exc_info=True)
             self._error_count += 1
@@ -179,3 +227,9 @@ class HATextAISensor(CoordinatorEntity, SensorEntity):
             self._state = STATE_ERROR
 
         self.async_write_ha_state()
+
+    async def async_reset_error_count(self) -> None:
+        """Reset the error counter."""
+        self._error_count = 0
+        self._last_error = None
+        self.async_write_ha_state()

custom_components/ha_text_ai/services.yaml

@@ -4,6 +4,7 @@ ask_question:
     Send a question to the AI model and receive a detailed response.
     The response will be stored in the conversation history and can be retrieved later.
     Response time may vary based on model selection and server load.
+    Supports various AI models with different capabilities and pricing.
   fields:
     question:
       name: Question
@@ -11,6 +12,7 @@ ask_question:
         Your question or prompt for the AI assistant. Be specific and clear for better results.
         You can ask about home automation, technical advice, or general questions.
         For complex queries, consider breaking them into smaller parts.
+        The system will maintain conversation context for follow-up questions.
       required: true
       example: |
         What automations would you recommend for a smart kitchen?
@@ -28,13 +30,17 @@ ask_question:
       name: Model
       description: >-
         Select an AI model to use (optional, overrides default setting).
-        Different models have different capabilities and token limits.
+        Different models have different capabilities, context limits, and response characteristics:
+        - GPT-3.5 Turbo: Fast, efficient, good for most tasks
+        - GPT-4: More capable, better reasoning, slower
+        - Claude-3: Advanced capabilities, longer context
         Note: More capable models may have longer response times and higher API costs.
       required: false
       example: "gpt-3.5-turbo"
       default: "gpt-3.5-turbo"
       selector:
         select:
+          custom_value: true
           options:
             - label: "GPT-3.5 Turbo (Fast & Efficient)"
               value: "gpt-3.5-turbo"
@@ -46,12 +52,16 @@ ask_question:
               value: "gpt-4-32k"
             - label: "GPT-4 Turbo (Latest)"
               value: "gpt-4-1106-preview"
+            - label: "Claude-3 Sonnet (Balanced)"
+              value: "claude-3-sonnet"
+            - label: "Claude-3 Opus (Most Advanced)"
+              value: "claude-3-opus"
           mode: dropdown
 
     temperature:
       name: Temperature
       description: >-
-        Controls response creativity (0-2):
+        Controls response creativity (0.0-2.0):
         0.0-0.3: Focused, consistent responses (best for technical/factual queries)
         0.4-0.7: Balanced responses (recommended for most uses)
         0.8-2.0: More creative, varied responses (best for brainstorming)
@@ -69,12 +79,12 @@ ask_question:
     max_tokens:
       name: Max Tokens
       description: >-
-        Maximum length of the response. Higher values allow longer responses but use more API tokens.
+        Maximum length of the response (1-4096 tokens).
         Recommended ranges:
-        - Short responses (256-512): Quick answers, status updates
-        - Medium responses (512-1024): Detailed explanations, instructions
-        - Long responses (1024-4096): Complex analysis, multiple examples
-        Note: Actual response length may be shorter based on content.
+        - Short (256-512): Quick answers, status updates
+        - Medium (512-1024): Detailed explanations
+        - Long (1024-4096): Complex analysis
+        Note: Token limits vary by model. Actual length may be shorter.
       required: false
       default: 1000
       selector:
@@ -84,19 +94,35 @@ ask_question:
           step: 1
           mode: box
 
+    system_prompt:
+      name: System Prompt
+      description: >-
+        Optional system prompt to set context for this specific question.
+        This will temporarily override the default system prompt.
+        Use this to specify expertise areas, response format, or special instructions.
+      required: false
+      example: "You are a home automation expert focused on energy efficiency and security"
+      selector:
+        text:
+          multiline: true
+
 clear_history:
   name: Clear History
   description: >-
     Delete all stored questions and responses from the conversation history.
     This action cannot be undone. Consider using 'get_history' first if you need to backup the data.
-    System prompt settings will be preserved.
+    System prompt settings and configuration will be preserved.
   fields: {}
 
 get_history:
   name: Get History
   description: >-
-    Retrieve recent conversation history, including questions, responses, and timestamps.
-    Results are ordered from newest to oldest and include metadata like model used and response times.
+    Retrieve recent conversation history, including:
+    - Questions and responses
+    - Timestamps and response times
+    - Models used and token counts
+    - Temperature and other settings
+    Results are ordered from newest to oldest.
   fields:
     limit:
       name: Limit
@@ -121,6 +147,7 @@ get_history:
       required: false
       selector:
         select:
+          custom_value: true
           options:
             - label: "All Models"
               value: ""
@@ -128,16 +155,41 @@ get_history:
               value: "gpt-3.5-turbo"
             - label: "GPT-4"
               value: "gpt-4"
+            - label: "GPT-4 Turbo"
+              value: "gpt-4-1106-preview"
+            - label: "Claude-3 Sonnet"
+              value: "claude-3-sonnet"
+            - label: "Claude-3 Opus"
+              value: "claude-3-opus"
           mode: dropdown
 
+    start_date:
+      name: Start Date
+      description: >-
+        Optional start date for filtering history.
+        Format: YYYY-MM-DD
+      required: false
+      selector:
+        datetime:
+          type: date
+
 set_system_prompt:
-  fields:  
+  name: Set System Prompt
+  description: >-
+    Set default system behavior instructions for all future conversations.
+    This defines how the AI should behave and respond to questions.
+    The prompt will be used for all models unless overridden per question.
+  fields:
     prompt:
       name: System Prompt
       description: >-
         Instructions that define how the AI should behave and respond.
-        Be specific about the desired expertise, tone, and format of responses.
-        Maximum length: 1000 characters.
+        Be specific about:
+        - Desired expertise and knowledge areas
+        - Response tone and style
+        - Output format preferences
+        - Special handling instructions
+        Maximum length: 1000 characters
       required: true
       example: |
         You are a home automation expert assistant. Focus on:
@@ -151,4 +203,3 @@ set_system_prompt:
       selector:
         text:
           multiline: true
-          type: text

custom_components/ha_text_ai/translations/en.json

@@ -3,26 +3,48 @@
         "step": {
             "user": {
                 "title": "Set up HA Text AI",
-                "description": "Configure your OpenAI integration",
+                "description": "Configure your AI integration for various providers (OpenAI, Anthropic, etc.). Different models have different capabilities and pricing.",
                 "data": {
-                    "api_key": "Your OpenAI API key",
-                    "model": "AI model to use for responses",
-                    "temperature": "Temperature for response generation (0-2)",
-                    "max_tokens": "Maximum tokens in response (1-4096)",
-                    "api_endpoint": "API endpoint URL",
-                    "request_interval": "Minimum time between API requests (seconds)"
+                    "api_key": "API key for authentication (required)",
+                    "model": "AI model to use (e.g., gpt-3.5-turbo, gpt-4, claude-3-sonnet, claude-3-opus)",
+                    "temperature": "Response creativity (0-2, lower = more focused and consistent)",
+                    "max_tokens": "Maximum response length (1-4096 tokens)",
+                    "api_endpoint": "Custom API endpoint URL (default varies by provider)",
+                    "request_interval": "Minimum time between requests in seconds (min: 0.1)",
+                    "system_prompt": "Default instructions for AI behavior and expertise"
                 }
             }
+        },
+        "error": {
+            "invalid_auth": "Authentication failed - check your API key",
+            "invalid_api_key": "Invalid API key - please verify your credentials",
+            "cannot_connect": "Connection failed - check endpoint and network status",
+            "invalid_model": "Model unavailable or not supported by the API",
+            "rate_limit": "Rate limit exceeded - please reduce request frequency",
+            "context_length": "Input exceeds maximum context length for model",
+            "api_error": "API service error - check provider status",
+            "timeout": "Request timeout - server not responding",
+            "queue_full": "Request queue full - try again later",
+            "invalid_prompt": "Invalid system prompt format or length",
+            "invalid_url_format": "Invalid API endpoint URL format",
+            "invalid_input": "Invalid configuration parameters",
+            "unknown": "Unexpected error - check logs for details"
         }
     },
     "options": {
         "step": {
             "init": {
-                "title": "HA Text AI Options",
+                "title": "HA Text AI Settings",
+                "description": "Adjust your AI integration parameters and behavior",
                 "data": {
-                    "temperature": "Response temperature (0-2)",
-                    "max_tokens": "Maximum response length",
-                    "request_interval": "Time between requests"
+                    "model": "Select AI model for responses (capabilities vary)",
+                    "temperature": "Response creativity (0-2, affects variation)",
+                    "max_tokens": "Maximum response length in tokens",
+                    "request_interval": "Minimum seconds between requests (rate limiting)",
+                    "system_prompt": "Default AI behavior and expertise instructions",
+                    "timeout": "Request timeout in seconds (default: 30)",
+                    "retry_count": "Number of retry attempts for failed requests",
+                    "queue_size": "Maximum pending requests in queue"
                 }
             }
         }
@@ -30,25 +52,96 @@
     "services": {
         "ask_question": {
             "name": "Ask Question",
-            "description": "Send a question to the AI model",
-            "fields": {  
+            "description": "Send a question to the AI model and get a detailed response. Supports context awareness for follow-up questions.",
+            "fields": {
                 "question": {
                     "name": "Question",
-                    "description": "Your question for the AI"
+                    "description": "Your question or prompt for the AI model. Be specific for better results."
+                },
+                "system_prompt": {
+                    "name": "System Prompt",
+                    "description": "Optional behavior instructions for this specific question"
+                },
+                "model": {
+                    "name": "Model",
+                    "description": "Optional specific AI model for this request (overrides default)"
+                },
+                "temperature": {
+                    "name": "Temperature",
+                    "description": "Optional creativity setting (0-2, affects response variation)"
+                },
+                "max_tokens": {
+                    "name": "Max Tokens",
+                    "description": "Optional maximum response length (1-4096 tokens)"
                 }
             }
         },
         "clear_history": {
             "name": "Clear History",
-            "description": "Clear conversation history"
+            "description": "Delete all stored conversation history, responses, and metadata"
         },
         "get_history": {
             "name": "Get History",
-            "description": "Retrieve conversation history"
+            "description": "Retrieve conversation history with detailed metadata including timestamps",
+            "fields": {
+                "limit": {
+                    "name": "Limit",
+                    "description": "Maximum number of conversations to return (1-100)"
+                },
+                "filter_model": {
+                    "name": "Filter Model",
+                    "description": "Show only responses from a specific model"
+                },
+                "start_date": {
+                    "name": "Start Date",
+                    "description": "Filter conversations from this date (YYYY-MM-DD)"
+                }
+            }
         },
         "set_system_prompt": {
             "name": "Set System Prompt",
-            "description": "Set system behavior instructions"
+            "description": "Update default AI behavior and expertise instructions",
+            "fields": {
+                "prompt": {
+                    "name": "System Prompt",
+                    "description": "Instructions for AI behavior, expertise, and response style"
+                }
+            }
+        }
+    },
+    "state": {
+        "ready": "Ready for requests",
+        "processing": "Processing request",
+        "error": "Error occurred",
+        "disconnected": "API disconnected",
+        "rate_limited": "Rate limit reached",
+        "initializing": "Starting up",
+        "retrying": "Retrying request",
+        "queued": "Request queued"
+    },
+    "entity": {
+        "sensor": {
+            "status": {
+                "name": "AI Status",
+                "state": {
+                    "ready": "Ready",
+                    "processing": "Processing",
+                    "error": "Error",
+                    "disconnected": "Disconnected",
+                    "rate_limited": "Rate Limited",
+                    "initializing": "Initializing",
+                    "retrying": "Retrying",
+                    "queued": "Queued"
+                }
+            },
+            "last_response": {
+                "name": "Last Response",
+                "state": {
+                    "success": "Success",
+                    "error": "Error",
+                    "timeout": "Timeout"
+                }
+            }
         }
     }
 }

custom_components/ha_text_ai/translations/ru.json

@@ -1,54 +0,0 @@
-{
-    "config": {
-        "step": {
-            "user": {
-                "title": "Настройка HA Text AI",
-                "description": "Настройка интеграции с OpenAI",
-                "data": {
-                    "api_key": "Ваш ключ API OpenAI",
-                    "model": "Модель ИИ для генерации ответов",
-                    "temperature": "Температура генерации ответов (0-2)",
-                    "max_tokens": "Максимальное количество токенов в ответе (1-4096)",
-                    "api_endpoint": "URL конечной точки API",
-                    "request_interval": "Минимальное время между запросами к API (секунды)"
-                }
-            }
-        }
-    },
-    "options": {
-        "step": {
-            "init": {
-                "title": "Настройки HA Text AI",
-                "data": {
-                    "temperature": "Температура ответов (0-2)",
-                    "max_tokens": "Максимальная длина ответа",
-                    "request_interval": "Время между запросами"
-                }
-            }
-        }
-    },
-    "services": {
-        "ask_question": {
-            "name": "Задать вопрос",
-            "description": "Отправить вопрос модели ИИ",
-            "fields": {
-                "question": {
-                    "name": "Вопрос",
-                    "description": "Ваш вопрос для ИИ"
-                }
-            }
-        },
-        "clear_history": {
-            "name": "Очистить историю",
-            "description": "Очистить историю разговора"
-        },
-        "get_history": {
-            "name": "Получить историю",
-            "description": "Получить историю разговора"
-        },
-        "set_system_prompt": {
-            "name": "Установить системный промпт",
-            "description": "Установить инструкции поведения системы"
-        }
-    }
-}

hacs.json

@@ -4,6 +4,6 @@
   "domains": ["sensor"],
   "homeassistant": "2024.11.0",
   "icon": "mdi:brain",
-  "version": "1.0.9",
+  "version": "2.0.0",
   "documentation": "https://github.com/smkrv/ha-text-ai"
 }