Support inbuilt functionality for controlling home-assistant

.github/settings.yml

@@ -10,7 +10,7 @@ repository:
   description: This custom component for Home Assistant allows you to generate text responses using GigaChain LLM framework (like GigaChat or YandexGPT and ChatGPT).
 
   # A URL with more information about the repository
-  homepage: https://github.com/gritaro/gigachat
+  homepage: https://github.com/gritaro/gigachain
 
   # A comma-separated list of topics to set on the repository
   topics: openai, gpt, homeassistant, voice-assistant, hacs-integration, chatgpt, yandexgpt, gigachat, gigachain, langchain

.github/workflows/hacs.yaml

@@ -0,0 +1,16 @@
+name: HACS Action
+
+on:
+  push:
+    branches:
+    - main
+
+jobs:
+  hacs:
+    name: HACS Action
+    runs-on: "ubuntu-latest"
+    steps:
+      - name: HACS Action
+        uses: "hacs/action@main"
+        with:
+          category: "integration"

.github/workflows/hassfest.yaml

@@ -0,0 +1,13 @@
+name: Validate with hassfest
+
+on:
+  push:
+    branches:
+    - main
+
+jobs:
+  validate:
+    runs-on: "ubuntu-latest"
+    steps:
+        - uses: "actions/checkout@v3"
+        - uses: "home-assistant/actions/hassfest@master"

.pre-commit-config.yaml

@@ -0,0 +1,15 @@
+repos:
+  - repo: https://github.com/charliermarsh/ruff-pre-commit
+    rev: v0.0.280
+    hooks:
+      - id: ruff
+
+  - repo: https://github.com/psf/black
+    rev: 23.7.0
+    hooks:
+      - id: black
+
+  - repo: https://github.com/PyCQA/isort
+    rev: 5.12.0
+    hooks:
+      - id: isort

README-ru.md

@@ -87,10 +87,20 @@
 Температура выборки. Значение температуры должно быть не меньше ноля. Чем выше значение, тем более случайным будет ответ модели. При значениях температуры больше двух, набор токенов в ответе модели может отличаться избыточной случайностью.
 Значение по умолчанию зависит от выбранной модели.
 
-* Максимум токенов (max_tokens, int)
+* Максимум токенов (max_tokens, `int`)
 
 Максимальное количество токенов, которые будут использованы для создания ответов.
 
+* _Использовать встроенный HA командный процессор_ (process_builtin_sentences, `bool`)
+
+Если включено, все фразы сначала будут отдаваться [встроенному в HA процессору шаблонных фраз](https://www.home-assistant.io/voice_control/builtin_sentences).
+Это основное поведение встроенной в Home Assistant диалоговой системы, что позволяет использовать команды вида `включи телевизор в зале`.
+Если фраза не может быть распознана встроенным процессором - она будет передана дальше, выбранной языковой модели.
+
+* История сообщений (chat_history, `bool`)
+
+Если у вашей модели дорогой тариф, либо ваш сценарий использования это позволяет, вы можете отключить историю. В противном случае вся история диалога передаётся в каждом запросе.
+
 ## Использование в качестве диалоговой системы
 Создайте и настройте новый голосовой ассистент:
 

README.md

@@ -82,10 +82,21 @@ Language model is used for text generation
 A value that determines the level of creativity and risk-taking the model should use when generating text. 
 A higher temperature means the model is more likely to generate unexpected results, while a lower temperature results in more deterministic results.
 
-* Max Tokens (max_tokens, int)
+* Max Tokens (max_tokens, `int`)
 
 The maximum number of words or “tokens” that the AI model should generate in its completion of the prompt.
 
+* _Process HA Builtin Sentences_ (process_builtin_sentences, `bool`)
+
+If enabled, integration first will pass all sentences to [HA built-in sentence processor](https://www.home-assistant.io/voice_control/builtin_sentences).
+This is default behaviour of default Home Assistant Voice Assistant engine which allow you to use commands something like `turn on the living room light`.
+If sentence will not be recognized by HA, it will be passed further to chosen LLM.
+
+* Chat History (chat_history, `bool`)
+
+Keep all conversation history. 
+
+
 ## Using as Voice Assistant
 Create and configure Voice Assistant:
 

custom_components/gigachain/__init__.py

@@ -3,20 +3,22 @@
 import logging
 from typing import Literal
 
+from home_assistant_intents import get_languages
 from homeassistant.components import conversation
 from homeassistant.components.conversation import agent
 from homeassistant.config_entries import ConfigEntry
-from homeassistant.const import MATCH_ALL
 from homeassistant.core import HomeAssistant
 from homeassistant.helpers import intent, template
 from homeassistant.util import ulid
-from langchain.schema import BaseMessage, HumanMessage, SystemMessage
+from langchain.schema import BaseMessage, HumanMessage, SystemMessage, AIMessage
 
 from .client_util import get_client
 from .const import (CONF_API_KEY, CONF_CHAT_MODEL, CONF_CHAT_MODEL_USER,
                     CONF_ENGINE, CONF_FOLDER_ID, CONF_MAX_TOKENS,
                     CONF_PROFANITY, CONF_PROMPT, CONF_TEMPERATURE,
                     DEFAULT_CHAT_MODEL, DEFAULT_PROFANITY, DEFAULT_PROMPT,
+                    CONF_PROCESS_BUILTIN_SENTENCES, DEFAULT_PROCESS_BUILTIN_SENTENCES,
+                    CONF_CHAT_HISTORY, DEFAULT_CHAT_HISTORY,
                     DEFAULT_TEMPERATURE, DOMAIN, ID_GIGACHAT)
 
 LOGGER = logging.getLogger(__name__)
@@ -50,7 +52,11 @@ async def async_setup_entry(hass: HomeAssistant, entry: ConfigEntry) -> bool:
     _client = await get_client(engine, common_args, entry)
 
     hass.data.setdefault(DOMAIN, {})[entry.entry_id] = _client
-    conversation.async_set_agent(hass, entry, GigaChatAI(hass, entry))
+    _agent = GigaChatAI(hass, entry)
+    await _agent.async_initialize(
+        hass.data.get("conversation_config")
+    )
+    conversation.async_set_agent(hass, entry, _agent)
     return True
 
 
@@ -61,31 +67,48 @@ async def async_unload_entry(hass: HomeAssistant, entry: ConfigEntry) -> bool:
     return True
 
 
-class GigaChatAI(conversation.AbstractConversationAgent):
+class GigaChatAI(conversation.DefaultAgent):
     def __init__(self, hass: HomeAssistant, entry: ConfigEntry) -> None:
         """Initialize the agent."""
+        super().__init__(hass)
         self.hass = hass
         self.entry = entry
         self.history: dict[str, list[BaseMessage]] = {}
 
     @property
     def supported_languages(self) -> list[str] | Literal["*"]:
         """Return a list of supported languages."""
-        return MATCH_ALL
+        return get_languages()
 
     async def async_process(
             self, user_input: agent.ConversationInput
     ) -> agent.ConversationResult:
         """Process a sentence."""
         raw_prompt = self.entry.options.get(CONF_PROMPT, DEFAULT_PROMPT)
-        if user_input.conversation_id in self.history:
+        chat_history_enabled = self.entry.options.get(CONF_CHAT_HISTORY, DEFAULT_CHAT_HISTORY)
+
+        if user_input.conversation_id in self.history and chat_history_enabled:
             conversation_id = user_input.conversation_id
             messages = self.history[conversation_id]
         else:
             conversation_id = ulid.ulid()
             prompt = self._async_generate_prompt(raw_prompt)
             messages = [SystemMessage(content=prompt)]
+
         messages.append(HumanMessage(content=user_input.text))
+
+        use_builtin_sentences = self.entry.options.get(CONF_PROCESS_BUILTIN_SENTENCES,
+                                                       DEFAULT_PROCESS_BUILTIN_SENTENCES)
+        if use_builtin_sentences:
+            default_agent_response = await super(GigaChatAI, self).async_process(user_input)
+
+            if default_agent_response.response.intent:
+                messages.append(AIMessage(content=default_agent_response.response.speech.get("plain").get("speech")))
+                self.history[conversation_id] = messages
+                return agent.ConversationResult(
+                    conversation_id=conversation_id, response=default_agent_response.response
+                )
+
         _client = self.hass.data[DOMAIN][self.entry.entry_id]
 
         try:
@@ -103,11 +126,11 @@ async def async_process(
 
         messages.append(res)
         self.history[conversation_id] = messages
-        LOGGER.debug(messages)
+        LOGGER.info(messages)
 
         response = intent.IntentResponse(language=user_input.language)
         response.async_set_speech(res.content)
-        LOGGER.debug(response)
+        LOGGER.info(response)
         return agent.ConversationResult(
             conversation_id=conversation_id, response=response
         )

custom_components/gigachain/config_flow.py

@@ -26,6 +26,8 @@
                     DEFAULT_MODELS, DEFAULT_PROFANITY, DEFAULT_PROMPT,
                     DEFAULT_SKIP_VALIDATION, DEFAULT_TEMPERATURE, DOMAIN,
                     ID_GIGACHAT, ID_OPENAI, ID_YANDEX_GPT, UNIQUE_ID,
+                    CONF_PROCESS_BUILTIN_SENTENCES, DEFAULT_PROCESS_BUILTIN_SENTENCES,
+                    CONF_CHAT_HISTORY, DEFAULT_CHAT_HISTORY,
                     UNIQUE_ID_GIGACHAT)
 
 LOGGER = logging.getLogger(__name__)
@@ -66,6 +68,7 @@
         CONF_PROMPT: DEFAULT_PROMPT,
         CONF_CHAT_MODEL: DEFAULT_CHAT_MODEL,
         CONF_CHAT_MODEL_USER: DEFAULT_CHAT_MODEL,
+        CONF_PROCESS_BUILTIN_SENTENCES: DEFAULT_PROCESS_BUILTIN_SENTENCES,
     }
 )
 
@@ -211,7 +214,19 @@ def common_config_option_schema(
             description={
                 "suggested_value": options.get(CONF_MAX_TOKENS)
             },
-        ): int
+        ): int,
+        vol.Optional(
+            CONF_PROCESS_BUILTIN_SENTENCES,
+            description={
+                "suggested_value": options.get(CONF_PROCESS_BUILTIN_SENTENCES, DEFAULT_PROCESS_BUILTIN_SENTENCES)
+            },
+            default=DEFAULT_PROCESS_BUILTIN_SENTENCES): bool,
+        vol.Optional(
+            CONF_CHAT_HISTORY,
+            description={
+                "suggested_value": options.get(CONF_CHAT_HISTORY, DEFAULT_CHAT_HISTORY)
+            },
+            default=DEFAULT_CHAT_HISTORY): bool
     })
     if unique_id == UNIQUE_ID_GIGACHAT:
         schema = schema.extend(

custom_components/gigachain/const.py

@@ -14,6 +14,10 @@
 CONF_MAX_TOKENS = "max_tokens"
 CONF_SKIP_VALIDATION = "skip_validation"
 DEFAULT_SKIP_VALIDATION = False
+CONF_PROCESS_BUILTIN_SENTENCES = "process_builtin_sentences"
+DEFAULT_PROCESS_BUILTIN_SENTENCES = True
+CONF_CHAT_HISTORY = "chat_history"
+DEFAULT_CHAT_HISTORY = True
 CONF_PROMPT = "prompt"
 DEFAULT_PROMPT = """Ты HAL 9000, компьютер из цикла произведений «Космическая одиссея» Артура Кларка, обладающий способностью к самообучению.
 Мы находимся в умном доме под управлением системы Home Assistant.

custom_components/gigachain/manifest.json

@@ -10,8 +10,9 @@
   "iot_class": "cloud_polling",
   "issue_tracker": "https://github.com/gritaro/gigachain/issues",
   "requirements": [
+    "home-assistant-intents",
     "gigachain==0.1.4",
     "yandexcloud==0.259.0"
   ],
-  "version": "0.1.4"
+  "version": "0.1.5"
 }

custom_components/gigachain/strings.json

@@ -54,7 +54,9 @@
           "model_openai": "Custom Model Name (leave empty to use from list above)",
           "temperature": "Temperature",
           "max_tokens": "Max Tokens",
-          "profanity": "Profanity"
+          "profanity": "Profanity",
+          "process_builtin_sentences": "Process HA Builtin Sentences",
+          "chat_history": "Chat History"
         }
       }
     }

custom_components/gigachain/translations/en.json

@@ -54,7 +54,9 @@
           "model_openai": "Custom Model Name (leave empty to use from list above)",
           "temperature": "Temperature",
           "max_tokens": "Max Tokens",
-          "profanity": "Profanity"
+          "profanity": "Profanity",
+          "process_builtin_sentences": "Process HA Builtin Sentences",
+          "chat_history": "Chat History"
         }
       }
     }

custom_components/gigachain/translations/ru.json

@@ -54,7 +54,9 @@
           "model_openai": "Своё имя модели (оставьте пустым для использования имени из списка)",
           "temperature": "Температура",
           "max_tokens": "Максимум токенов",
-          "profanity": "Цензура"
+          "profanity": "Цензура",
+          "process_builtin_sentences": "Использовать встроенный HA командный процессор",
+          "chat_history": "История сообщений"
         }
       }
     }