dify

dify/api/core/tools/builtin_tool/_position.yaml

@@ -0,0 +1,4 @@
+- audio
+- code
+- time
+- webscraper

dify/api/core/tools/builtin_tool/provider.py

@@ -0,0 +1,221 @@
+from abc import abstractmethod
+from os import listdir, path
+from typing import Any
+
+from core.entities.provider_entities import ProviderConfig
+from core.helper.module_import_helper import load_single_subclass_from_source
+from core.plugin.entities.plugin_daemon import CredentialType
+from core.tools.__base.tool_provider import ToolProviderController
+from core.tools.__base.tool_runtime import ToolRuntime
+from core.tools.builtin_tool.tool import BuiltinTool
+from core.tools.entities.tool_entities import (
+    OAuthSchema,
+    ToolEntity,
+    ToolProviderEntity,
+    ToolProviderType,
+)
+from core.tools.entities.values import ToolLabelEnum, default_tool_label_dict
+from core.tools.errors import (
+    ToolProviderNotFoundError,
+)
+from core.tools.utils.yaml_utils import load_yaml_file_cached
+
+
+class BuiltinToolProviderController(ToolProviderController):
+    tools: list[BuiltinTool]
+
+    def __init__(self, **data: Any):
+        self.tools = []
+
+        # load provider yaml
+        provider = self.__class__.__module__.split(".")[-1]
+        yaml_path = path.join(path.dirname(path.realpath(__file__)), "providers", provider, f"{provider}.yaml")
+        try:
+            provider_yaml = load_yaml_file_cached(yaml_path)
+        except Exception as e:
+            raise ToolProviderNotFoundError(f"can not load provider yaml for {provider}: {e}")
+
+        if "credentials_for_provider" in provider_yaml and provider_yaml["credentials_for_provider"] is not None:
+            # set credentials name
+            for credential_name in provider_yaml["credentials_for_provider"]:
+                provider_yaml["credentials_for_provider"][credential_name]["name"] = credential_name
+
+        credentials_schema = []
+        for credential in provider_yaml.get("credentials_for_provider", {}):
+            credential_dict = provider_yaml.get("credentials_for_provider", {}).get(credential, {})
+            credentials_schema.append(credential_dict)
+
+        oauth_schema = None
+        if provider_yaml.get("oauth_schema", None) is not None:
+            oauth_schema = OAuthSchema(
+                client_schema=provider_yaml.get("oauth_schema", {}).get("client_schema", []),
+                credentials_schema=provider_yaml.get("oauth_schema", {}).get("credentials_schema", []),
+            )
+
+        super().__init__(
+            entity=ToolProviderEntity(
+                identity=provider_yaml["identity"],
+                credentials_schema=credentials_schema,
+                oauth_schema=oauth_schema,
+            ),
+        )
+
+        self._load_tools()
+
+    def _load_tools(self):
+        provider = self.entity.identity.name
+        tool_path = path.join(path.dirname(path.realpath(__file__)), "providers", provider, "tools")
+        # get all the yaml files in the tool path
+        tool_files = list(filter(lambda x: x.endswith(".yaml") and not x.startswith("__"), listdir(tool_path)))
+        tools = []
+        for tool_file in tool_files:
+            # get tool name
+            tool_name = tool_file.split(".")[0]
+            tool = load_yaml_file_cached(path.join(tool_path, tool_file))
+
+            # get tool class, import the module
+            assistant_tool_class: type = load_single_subclass_from_source(
+                module_name=f"core.tools.builtin_tool.providers.{provider}.tools.{tool_name}",
+                script_path=path.join(
+                    path.dirname(path.realpath(__file__)),
+                    "builtin_tool",
+                    "providers",
+                    provider,
+                    "tools",
+                    f"{tool_name}.py",
+                ),
+                parent_type=BuiltinTool,
+            )
+            tool["identity"]["provider"] = provider
+            tools.append(
+                assistant_tool_class(
+                    provider=provider,
+                    entity=ToolEntity.model_validate(tool),
+                    runtime=ToolRuntime(tenant_id=""),
+                )
+            )
+
+        self.tools = tools
+
+    def _get_builtin_tools(self) -> list[BuiltinTool]:
+        """
+        returns a list of tools that the provider can provide
+
+        :return: list of tools
+        """
+        return self.tools
+
+    def get_credentials_schema(self) -> list[ProviderConfig]:
+        """
+        returns the credentials schema of the provider
+
+        :return: the credentials schema
+        """
+        return self.get_credentials_schema_by_type(CredentialType.API_KEY)
+
+    def get_credentials_schema_by_type(self, credential_type: str) -> list[ProviderConfig]:
+        """
+        returns the credentials schema of the provider
+
+        :param credential_type: the type of the credential
+        :return: the credentials schema of the provider
+        """
+        if credential_type == CredentialType.OAUTH2.value:
+            return self.entity.oauth_schema.credentials_schema.copy() if self.entity.oauth_schema else []
+        if credential_type == CredentialType.API_KEY:
+            return self.entity.credentials_schema.copy() if self.entity.credentials_schema else []
+        raise ValueError(f"Invalid credential type: {credential_type}")
+
+    def get_oauth_client_schema(self) -> list[ProviderConfig]:
+        """
+        returns the oauth client schema of the provider
+
+        :return: the oauth client schema
+        """
+        return self.entity.oauth_schema.client_schema.copy() if self.entity.oauth_schema else []
+
+    def get_supported_credential_types(self) -> list[CredentialType]:
+        """
+        returns the credential support type of the provider
+        """
+        types = []
+        if self.entity.credentials_schema is not None and len(self.entity.credentials_schema) > 0:
+            types.append(CredentialType.API_KEY)
+        if self.entity.oauth_schema is not None and len(self.entity.oauth_schema.credentials_schema) > 0:
+            types.append(CredentialType.OAUTH2)
+        return types
+
+    def get_tools(self) -> list[BuiltinTool]:
+        """
+        returns a list of tools that the provider can provide
+
+        :return: list of tools
+        """
+        return self._get_builtin_tools()
+
+    def get_tool(self, tool_name: str) -> BuiltinTool | None:  # type: ignore
+        """
+        returns the tool that the provider can provide
+        """
+        return next(filter(lambda x: x.entity.identity.name == tool_name, self.get_tools()), None)
+
+    @property
+    def need_credentials(self) -> bool:
+        """
+        returns whether the provider needs credentials
+
+        :return: whether the provider needs credentials
+        """
+        return (
+            self.entity.credentials_schema is not None
+            and len(self.entity.credentials_schema) != 0
+            or (self.entity.oauth_schema is not None and len(self.entity.oauth_schema.credentials_schema) != 0)
+        )
+
+    @property
+    def provider_type(self) -> ToolProviderType:
+        """
+        returns the type of the provider
+
+        :return: type of the provider
+        """
+        return ToolProviderType.BUILT_IN
+
+    @property
+    def tool_labels(self) -> list[str]:
+        """
+        returns the labels of the provider
+
+        :return: labels of the provider
+        """
+        label_enums = self._get_tool_labels()
+        return [default_tool_label_dict[label].name for label in label_enums]
+
+    def _get_tool_labels(self) -> list[ToolLabelEnum]:
+        """
+        returns the labels of the provider
+        """
+        return self.entity.identity.tags or []
+
+    def validate_credentials(self, user_id: str, credentials: dict[str, Any]):
+        """
+        validate the credentials of the provider
+
+        :param user_id: use id
+        :param credentials: the credentials of the tool
+        """
+        # validate credentials format
+        self.validate_credentials_format(credentials)
+
+        # validate credentials
+        self._validate_credentials(user_id, credentials)
+
+    @abstractmethod
+    def _validate_credentials(self, user_id: str, credentials: dict[str, Any]):
+        """
+        validate the credentials of the provider
+
+        :param user_id: use id
+        :param credentials: the credentials of the tool
+        """
+        pass

dify/api/core/tools/builtin_tool/providers/_positions.py

@@ -0,0 +1,20 @@
+import os.path
+
+from core.helper.position_helper import get_tool_position_map, sort_by_position_map
+from core.tools.entities.api_entities import ToolProviderApiEntity
+
+
+class BuiltinToolProviderSort:
+    _position: dict[str, int] = {}
+
+    @classmethod
+    def sort(cls, providers: list[ToolProviderApiEntity]) -> list[ToolProviderApiEntity]:
+        if not cls._position:
+            cls._position = get_tool_position_map(os.path.join(os.path.dirname(__file__), ".."))
+
+        def name_func(provider: ToolProviderApiEntity) -> str:
+            return provider.name
+
+        sorted_providers = sort_by_position_map(cls._position, providers, name_func)
+
+        return sorted_providers

dify/api/core/tools/builtin_tool/providers/audio/_assets/icon.svg

@@ -0,0 +1,3 @@
+<svg xmlns="http://www.w3.org/2000/svg" width="200" height="200" viewBox="0 0 200 200" fill="none">
+<path d="M167.358 102.395C167.358 117.174 157.246 129.18 144.61 131.027H137.861C125.225 129.18 115.113 117.174 115.113 102.395H100.792C100.792 123.637 115.118 142.106 133.653 145.801V164.276H147.139V145.801C165.674 142.106 180 124.558 180 102.4H167.358V102.395ZM154.717 62.677C154.717 53.4397 147.979 46.9765 140.396 46.9765C138.523 46.9446 136.663 47.3273 134.924 48.1024C133.185 48.8775 131.603 50.0294 130.27 51.4909C128.936 52.9524 127.878 54.6943 127.157 56.6148C126.436 58.5354 126.066 60.5962 126.07 62.677V78.3775H154.717V70.4478V62.677ZM126.07 102.395C126.07 111.632 132.813 118.095 140.396 118.095C142.269 118.127 144.13 117.744 145.868 116.969C147.607 116.194 149.189 115.042 150.523 113.581C151.856 112.119 152.914 110.377 153.635 108.457C154.356 106.536 154.726 104.475 154.722 102.395V86.694H126.07V102.395ZM92.1297 45.8938L70.4796 21.7595L69.4235 20.5865L59.604 20L68.3674 20.5865L67.3113 21.7654L64.1429 25.2961L63.6149 25.8826L64.1429 27.0614L66.2552 29.4133L77.8723 42.3631H54.1099C35.1 43.5361 20.3146 61.1896 20.3146 81.7874V83.5527H28.2354V81.7932C28.2354 65.8992 39.8525 52.3628 54.1099 51.1899H77.8723L66.2552 64.1338L64.671 65.8992L64.1429 67.0722L63.6149 67.6645L64.1429 68.251L68.3674 72.9606L68.8954 73.5471L69.4235 72.9606L74.1759 67.6645L92.1297 47.6591L92.6578 47.0727L92.1297 45.8938ZM20 95.8496V118.213H30.033V107.034H50.099V168.821H40.066V180H70.165V168.821H60.132V107.034H80.198V118.213H90.231V95.8496H20Z" fill="#FF0099"/>
+</svg>

dify/api/core/tools/builtin_tool/providers/audio/audio.py

@@ -0,0 +1,8 @@
+from typing import Any
+
+from core.tools.builtin_tool.provider import BuiltinToolProviderController
+
+
+class AudioToolProvider(BuiltinToolProviderController):
+    def _validate_credentials(self, user_id: str, credentials: dict[str, Any]):
+        pass

dify/api/core/tools/builtin_tool/providers/audio/audio.yaml

@@ -0,0 +1,11 @@
+identity:
+  author: hjlarry
+  name: audio
+  label:
+    en_US: Audio
+  description:
+    en_US: A tool for tts and asr.
+    zh_Hans: 一个用于文本转语音和语音转文本的工具。
+  icon: icon.svg
+  tags:
+    - utilities

dify/api/core/tools/builtin_tool/providers/audio/tools/asr.py

@@ -0,0 +1,84 @@
+import io
+from collections.abc import Generator
+from typing import Any
+
+from core.file.enums import FileType
+from core.file.file_manager import download
+from core.model_manager import ModelManager
+from core.model_runtime.entities.model_entities import ModelType
+from core.plugin.entities.parameters import PluginParameterOption
+from core.tools.builtin_tool.tool import BuiltinTool
+from core.tools.entities.common_entities import I18nObject
+from core.tools.entities.tool_entities import ToolInvokeMessage, ToolParameter
+from services.model_provider_service import ModelProviderService
+
+
+class ASRTool(BuiltinTool):
+    def _invoke(
+        self,
+        user_id: str,
+        tool_parameters: dict[str, Any],
+        conversation_id: str | None = None,
+        app_id: str | None = None,
+        message_id: str | None = None,
+    ) -> Generator[ToolInvokeMessage, None, None]:
+        file = tool_parameters.get("audio_file")
+        if file.type != FileType.AUDIO:  # type: ignore
+            yield self.create_text_message("not a valid audio file")
+            return
+        audio_binary = io.BytesIO(download(file))  # type: ignore
+        audio_binary.name = "temp.mp3"
+        provider, model = tool_parameters.get("model").split("#")  # type: ignore
+        model_manager = ModelManager()
+        model_instance = model_manager.get_model_instance(
+            tenant_id=self.runtime.tenant_id,
+            provider=provider,
+            model_type=ModelType.SPEECH2TEXT,
+            model=model,
+        )
+        text = model_instance.invoke_speech2text(
+            file=audio_binary,
+            user=user_id,
+        )
+        yield self.create_text_message(text)
+
+    def get_available_models(self) -> list[tuple[str, str]]:
+        model_provider_service = ModelProviderService()
+        models = model_provider_service.get_models_by_model_type(
+            tenant_id=self.runtime.tenant_id, model_type="speech2text"
+        )
+        items = []
+        for provider_model in models:
+            provider = provider_model.provider
+            for model in provider_model.models:
+                items.append((provider, model.model))
+        return items
+
+    def get_runtime_parameters(
+        self,
+        conversation_id: str | None = None,
+        app_id: str | None = None,
+        message_id: str | None = None,
+    ) -> list[ToolParameter]:
+        parameters = []
+
+        options = []
+        for provider, model in self.get_available_models():
+            option = PluginParameterOption(value=f"{provider}#{model}", label=I18nObject(en_US=f"{model}({provider})"))
+            options.append(option)
+
+        parameters.append(
+            ToolParameter(
+                name="model",
+                label=I18nObject(en_US="Model", zh_Hans="Model"),
+                human_description=I18nObject(
+                    en_US="All available ASR models. You can config model in the Model Provider of Settings.",
+                    zh_Hans="所有可用的 ASR 模型。你可以在设置中的模型供应商里配置。",
+                ),
+                type=ToolParameter.ToolParameterType.SELECT,
+                form=ToolParameter.ToolParameterForm.FORM,
+                required=True,
+                options=options,
+            )
+        )
+        return parameters

dify/api/core/tools/builtin_tool/providers/audio/tools/asr.yaml

@@ -0,0 +1,22 @@
+identity:
+  name: asr
+  author: hjlarry
+  label:
+    en_US: Speech To Text
+description:
+  human:
+    en_US: Convert audio file to text.
+    zh_Hans: 将音频文件转换为文本。
+  llm: Convert audio file to text.
+parameters:
+  - name: audio_file
+    type: file
+    required: true
+    label:
+      en_US: Audio File
+      zh_Hans: 音频文件
+    human_description:
+      en_US: The audio file to be converted.
+      zh_Hans: 要转换的音频文件。
+    llm_description: The audio file to be converted.
+    form: llm

dify/api/core/tools/builtin_tool/providers/audio/tools/tts.py

@@ -0,0 +1,116 @@
+import io
+from collections.abc import Generator
+from typing import Any
+
+from core.model_manager import ModelManager
+from core.model_runtime.entities.model_entities import ModelPropertyKey, ModelType
+from core.plugin.entities.parameters import PluginParameterOption
+from core.tools.builtin_tool.tool import BuiltinTool
+from core.tools.entities.common_entities import I18nObject
+from core.tools.entities.tool_entities import ToolInvokeMessage, ToolParameter
+from services.model_provider_service import ModelProviderService
+
+
+class TTSTool(BuiltinTool):
+    def _invoke(
+        self,
+        user_id: str,
+        tool_parameters: dict[str, Any],
+        conversation_id: str | None = None,
+        app_id: str | None = None,
+        message_id: str | None = None,
+    ) -> Generator[ToolInvokeMessage, None, None]:
+        provider, model = tool_parameters.get("model").split("#")  # type: ignore
+        voice = tool_parameters.get(f"voice#{provider}#{model}")
+        model_manager = ModelManager()
+        if not self.runtime:
+            raise ValueError("Runtime is required")
+        model_instance = model_manager.get_model_instance(
+            tenant_id=self.runtime.tenant_id or "",
+            provider=provider,
+            model_type=ModelType.TTS,
+            model=model,
+        )
+        if not voice:
+            voices = model_instance.get_tts_voices()
+            if voices:
+                voice = voices[0].get("value")
+                if not voice:
+                    raise ValueError("Sorry, no voice available.")
+            else:
+                raise ValueError("Sorry, no voice available.")
+        tts = model_instance.invoke_tts(
+            content_text=tool_parameters.get("text"),  # type: ignore
+            user=user_id,
+            tenant_id=self.runtime.tenant_id,
+            voice=voice,
+        )
+        buffer = io.BytesIO()
+        for chunk in tts:
+            buffer.write(chunk)
+
+        wav_bytes = buffer.getvalue()
+        yield self.create_text_message("Audio generated successfully")
+        yield self.create_blob_message(
+            blob=wav_bytes,
+            meta={"mime_type": "audio/x-wav"},
+        )
+
+    def get_available_models(self) -> list[tuple[str, str, list[Any]]]:
+        if not self.runtime:
+            raise ValueError("Runtime is required")
+        model_provider_service = ModelProviderService()
+        tid: str = self.runtime.tenant_id or ""
+        models = model_provider_service.get_models_by_model_type(tenant_id=tid, model_type="tts")
+        items = []
+        for provider_model in models:
+            provider = provider_model.provider
+            for model in provider_model.models:
+                voices = model.model_properties.get(ModelPropertyKey.VOICES, [])
+                items.append((provider, model.model, voices))
+        return items
+
+    def get_runtime_parameters(
+        self,
+        conversation_id: str | None = None,
+        app_id: str | None = None,
+        message_id: str | None = None,
+    ) -> list[ToolParameter]:
+        parameters = []
+
+        options = []
+        for provider, model, voices in self.get_available_models():
+            option = PluginParameterOption(value=f"{provider}#{model}", label=I18nObject(en_US=f"{model}({provider})"))
+            options.append(option)
+            parameters.append(
+                ToolParameter(
+                    name=f"voice#{provider}#{model}",
+                    label=I18nObject(en_US=f"Voice of {model}({provider})"),
+                    human_description=I18nObject(en_US=f"Select a voice for {model} model"),
+                    placeholder=I18nObject(en_US="Select a voice"),
+                    type=ToolParameter.ToolParameterType.SELECT,
+                    form=ToolParameter.ToolParameterForm.FORM,
+                    options=[
+                        PluginParameterOption(value=voice.get("mode"), label=I18nObject(en_US=voice.get("name")))
+                        for voice in voices
+                    ],
+                )
+            )
+
+        parameters.insert(
+            0,
+            ToolParameter(
+                name="model",
+                label=I18nObject(en_US="Model", zh_Hans="Model"),
+                human_description=I18nObject(
+                    en_US="All available TTS models. You can config model in the Model Provider of Settings.",
+                    zh_Hans="所有可用的 TTS 模型。你可以在设置中的模型供应商里配置。",
+                ),
+                type=ToolParameter.ToolParameterType.SELECT,
+                form=ToolParameter.ToolParameterForm.FORM,
+                required=True,
+                placeholder=I18nObject(en_US="Select a model", zh_Hans="选择模型"),
+                options=options,
+            ),
+        )
+        return parameters

dify/api/core/tools/builtin_tool/providers/audio/tools/tts.yaml

@@ -0,0 +1,22 @@
+identity:
+  name: tts
+  author: hjlarry
+  label:
+    en_US: Text To Speech
+description:
+  human:
+    en_US: Convert text to audio file.
+    zh_Hans: 将文本转换为音频文件。
+  llm: Convert text to audio file.
+parameters:
+  - name: text
+    type: string
+    required: true
+    label:
+      en_US: Text
+      zh_Hans: 文本
+    human_description:
+      en_US: The text to be converted.
+      zh_Hans: 要转换的文本。
+    llm_description: The text to be converted.
+    form: llm

dify/api/core/tools/builtin_tool/providers/code/_assets/icon.svg

@@ -0,0 +1 @@
+<svg width="14" height="14" viewBox="0 0 14 14" fill="none" xmlns="http://www.w3.org/2000/svg" class="w-3.5 h-3.5" data-icon="Code" aria-hidden="true"><g id="icons/code"><path id="Vector (Stroke)" fill-rule="evenodd" clip-rule="evenodd" d="M8.32593 1.69675C8.67754 1.78466 8.89132 2.14096 8.80342 2.49257L6.47009 11.8259C6.38218 12.1775 6.02588 12.3913 5.67427 12.3034C5.32265 12.2155 5.10887 11.8592 5.19678 11.5076L7.53011 2.17424C7.61801 1.82263 7.97431 1.60885 8.32593 1.69675ZM3.96414 4.20273C4.22042 4.45901 4.22042 4.87453 3.96413 5.13081L2.45578 6.63914C2.45577 6.63915 2.45578 6.63914 2.45578 6.63914C2.25645 6.83851 2.25643 7.16168 2.45575 7.36103C2.45574 7.36103 2.45576 7.36104 2.45575 7.36103L3.96413 8.86936C4.22041 9.12564 4.22042 9.54115 3.96414 9.79744C3.70787 10.0537 3.29235 10.0537 3.03607 9.79745L1.52769 8.28913C0.815811 7.57721 0.815803 6.42302 1.52766 5.7111L3.03606 4.20272C3.29234 3.94644 3.70786 3.94644 3.96414 4.20273ZM10.0361 4.20273C10.2923 3.94644 10.7078 3.94644 10.9641 4.20272L12.4725 5.71108C13.1843 6.423 13.1844 7.57717 12.4725 8.28909L10.9641 9.79745C10.7078 10.0537 10.2923 10.0537 10.036 9.79744C9.77977 9.54115 9.77978 9.12564 10.0361 8.86936L11.5444 7.36107C11.7437 7.16172 11.7438 6.83854 11.5444 6.63917C11.5444 6.63915 11.5445 6.63918 11.5444 6.63917L10.0361 5.13081C9.77978 4.87453 9.77978 4.45901 10.0361 4.20273Z" fill="#2e90fa"></path></g></svg>

dify/api/core/tools/builtin_tool/providers/code/code.py

@@ -0,0 +1,8 @@
+from typing import Any
+
+from core.tools.builtin_tool.provider import BuiltinToolProviderController
+
+
+class CodeToolProvider(BuiltinToolProviderController):
+    def _validate_credentials(self, user_id: str, credentials: dict[str, Any]):
+        pass

dify/api/core/tools/builtin_tool/providers/code/code.yaml

@@ -0,0 +1,14 @@
+identity:
+  author: Dify
+  name: code
+  label:
+    en_US: Code Interpreter
+    zh_Hans: 代码解释器
+    pt_BR: Interpretador de Código
+  description:
+    en_US: Run a piece of code and get the result back.
+    zh_Hans: 运行一段代码并返回结果。
+    pt_BR: Execute um trecho de código e obtenha o resultado de volta.
+  icon: icon.svg
+  tags:
+    - productivity

dify/api/core/tools/builtin_tool/providers/code/tools/simple_code.py

@@ -0,0 +1,33 @@
+from collections.abc import Generator
+from typing import Any
+
+from core.helper.code_executor.code_executor import CodeExecutor, CodeLanguage
+from core.tools.builtin_tool.tool import BuiltinTool
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.errors import ToolInvokeError
+
+
+class SimpleCode(BuiltinTool):
+    def _invoke(
+        self,
+        user_id: str,
+        tool_parameters: dict[str, Any],
+        conversation_id: str | None = None,
+        app_id: str | None = None,
+        message_id: str | None = None,
+    ) -> Generator[ToolInvokeMessage, None, None]:
+        """
+        invoke simple code
+        """
+
+        language = tool_parameters.get("language", CodeLanguage.PYTHON3)
+        code = tool_parameters.get("code", "")
+
+        if language not in {CodeLanguage.PYTHON3, CodeLanguage.JAVASCRIPT}:
+            raise ValueError(f"Only python3 and javascript are supported, not {language}")
+
+        try:
+            result = CodeExecutor.execute_code(language, "", code)
+            yield self.create_text_message(result)
+        except Exception as e:
+            raise ToolInvokeError(str(e))

dify/api/core/tools/builtin_tool/providers/code/tools/simple_code.yaml

@@ -0,0 +1,51 @@
+identity:
+  name: simple_code
+  author: Dify
+  label:
+    en_US: Code Interpreter
+    zh_Hans: 代码解释器
+    pt_BR: Interpretador de Código
+description:
+  human:
+    en_US: Run code and get the result back. When you're using a lower quality model, please make sure there are some tips help LLM to understand how to write the code.
+    zh_Hans: 运行一段代码并返回结果。当您使用较低质量的模型时，请确保有一些提示帮助 LLM 理解如何编写代码。
+    pt_BR: Execute um trecho de código e obtenha o resultado de volta. quando você estiver usando um modelo de qualidade inferior, certifique-se de que existam algumas dicas para ajudar o LLM a entender como escrever o código.
+  llm: A tool for running code and getting the result back. Only native packages are allowed, network/IO operations are disabled. and you must use print() or console.log() to output the result or result will be empty.
+parameters:
+  - name: language
+    type: string
+    required: true
+    label:
+      en_US: Language
+      zh_Hans: 语言
+      pt_BR: Idioma
+    human_description:
+      en_US: The programming language of the code
+      zh_Hans: 代码的编程语言
+      pt_BR: A linguagem de programação do código
+    llm_description: language of the code, only "python3" and "javascript" are supported
+    form: llm
+    options:
+      - value: python3
+        label:
+          en_US: Python3
+          zh_Hans: Python3
+          pt_BR: Python3
+      - value: javascript
+        label:
+          en_US: JavaScript
+          zh_Hans: JavaScript
+          pt_BR: JavaScript
+  - name: code
+    type: string
+    required: true
+    label:
+      en_US: Code
+      zh_Hans: 代码
+      pt_BR: Código
+    human_description:
+      en_US: The code to be executed
+      zh_Hans: 要执行的代码
+      pt_BR: O código a ser executado
+    llm_description: code to be executed, only native packages are allowed, network/IO operations are disabled.
+    form: llm

dify/api/core/tools/builtin_tool/providers/time/_assets/icon.svg

@@ -0,0 +1,3 @@
+<svg xmlns="http://www.w3.org/2000/svg" width="16" height="16" viewBox="0 0 16 16" fill="none">
+  <path fill-rule="evenodd" clip-rule="evenodd" d="M0.666992 8.00008C0.666992 3.94999 3.95024 0.666748 8.00033 0.666748C12.0504 0.666748 15.3337 3.94999 15.3337 8.00008C15.3337 12.0502 12.0504 15.3334 8.00033 15.3334C3.95024 15.3334 0.666992 12.0502 0.666992 8.00008ZM8.66699 4.00008C8.66699 3.63189 8.36852 3.33341 8.00033 3.33341C7.63213 3.33341 7.33366 3.63189 7.33366 4.00008V8.00008C7.33366 8.2526 7.47633 8.48344 7.70218 8.59637L10.3688 9.9297C10.6982 10.0944 11.0986 9.96088 11.2633 9.63156C11.4279 9.30224 11.2945 8.90179 10.9651 8.73713L8.66699 7.58806V4.00008Z" fill="#EC4A0A"/>
+</svg>

dify/api/core/tools/builtin_tool/providers/time/time.py

@@ -0,0 +1,8 @@
+from typing import Any
+
+from core.tools.builtin_tool.provider import BuiltinToolProviderController
+
+
+class WikiPediaProvider(BuiltinToolProviderController):
+    def _validate_credentials(self, user_id: str, credentials: dict[str, Any]):
+        pass

dify/api/core/tools/builtin_tool/providers/time/time.yaml

@@ -0,0 +1,14 @@
+identity:
+  author: Dify
+  name: time
+  label:
+    en_US: CurrentTime
+    zh_Hans: 时间
+    pt_BR: CurrentTime
+  description:
+    en_US: A tool for getting the current time.
+    zh_Hans: 一个用于获取当前时间的工具。
+    pt_BR: A tool for getting the current time.
+  icon: icon.svg
+  tags:
+    - utilities

dify/api/core/tools/builtin_tool/providers/time/tools/current_time.py

@@ -0,0 +1,35 @@
+from collections.abc import Generator
+from datetime import UTC, datetime
+from typing import Any
+
+from pytz import timezone as pytz_timezone
+
+from core.tools.builtin_tool.tool import BuiltinTool
+from core.tools.entities.tool_entities import ToolInvokeMessage
+
+
+class CurrentTimeTool(BuiltinTool):
+    def _invoke(
+        self,
+        user_id: str,
+        tool_parameters: dict[str, Any],
+        conversation_id: str | None = None,
+        app_id: str | None = None,
+        message_id: str | None = None,
+    ) -> Generator[ToolInvokeMessage, None, None]:
+        """
+        invoke tools
+        """
+        # get timezone
+        tz = tool_parameters.get("timezone", "UTC")
+        fm = tool_parameters.get("format") or "%Y-%m-%d %H:%M:%S %Z"
+        if tz == "UTC":
+            yield self.create_text_message(f"{datetime.now(UTC).strftime(fm)}")
+            return
+
+        try:
+            tz = pytz_timezone(tz)
+        except Exception:
+            yield self.create_text_message(f"Invalid timezone: {tz}")
+            return
+        yield self.create_text_message(f"{datetime.now(tz).strftime(fm)}")

dify/api/core/tools/builtin_tool/providers/time/tools/current_time.yaml

@@ -0,0 +1,131 @@
+identity:
+  name: current_time
+  author: Dify
+  label:
+    en_US: Current Time
+    zh_Hans: 获取当前时间
+    pt_BR: Current Time
+description:
+  human:
+    en_US: A tool for getting the current time.
+    zh_Hans: 一个用于获取当前时间的工具。
+    pt_BR: A tool for getting the current time.
+  llm: A tool for getting the current time.
+parameters:
+  - name: format
+    type: string
+    required: false
+    label:
+      en_US: Format
+      zh_Hans: 格式
+      pt_BR: Format
+    human_description:
+      en_US: Time format in strftime standard.
+      zh_Hans: strftime 标准的时间格式。
+      pt_BR: Time format in strftime standard.
+    form: form
+    default: "%Y-%m-%d %H:%M:%S"
+  - name: timezone
+    type: select
+    required: false
+    label:
+      en_US: Timezone
+      zh_Hans: 时区
+      pt_BR: Timezone
+    human_description:
+      en_US: Timezone
+      zh_Hans: 时区
+      pt_BR: Timezone
+    form: form
+    default: UTC
+    options:
+      - value: UTC
+        label:
+          en_US: UTC
+          zh_Hans: UTC
+          pt_BR: UTC
+      - value: America/New_York
+        label:
+          en_US: America/New_York
+          zh_Hans: 美洲/纽约
+          pt_BR: America/New_York
+      - value: America/Los_Angeles
+        label:
+          en_US: America/Los_Angeles
+          zh_Hans: 美洲/洛杉矶
+          pt_BR: America/Los_Angeles
+      - value: America/Chicago
+        label:
+          en_US: America/Chicago
+          zh_Hans: 美洲/芝加哥
+          pt_BR: America/Chicago
+      - value: America/Sao_Paulo
+        label:
+          en_US: America/Sao_Paulo
+          zh_Hans: 美洲/圣保罗
+          pt_BR: América/São Paulo
+      - value: Asia/Shanghai
+        label:
+          en_US: Asia/Shanghai
+          zh_Hans: 亚洲/上海
+          pt_BR: Asia/Shanghai
+      - value: Asia/Ho_Chi_Minh
+        label:
+          en_US: Asia/Ho_Chi_Minh
+          zh_Hans: 亚洲/胡志明市
+          pt_BR: Ásia/Ho Chi Minh
+      - value: Asia/Tokyo
+        label:
+          en_US: Asia/Tokyo
+          zh_Hans: 亚洲/东京
+          pt_BR: Asia/Tokyo
+      - value: Asia/Dubai
+        label:
+          en_US: Asia/Dubai
+          zh_Hans: 亚洲/迪拜
+          pt_BR: Asia/Dubai
+      - value: Asia/Kolkata
+        label:
+          en_US: Asia/Kolkata
+          zh_Hans: 亚洲/加尔各答
+          pt_BR: Asia/Kolkata
+      - value: Asia/Seoul
+        label:
+          en_US: Asia/Seoul
+          zh_Hans: 亚洲/首尔
+          pt_BR: Asia/Seoul
+      - value: Asia/Singapore
+        label:
+          en_US: Asia/Singapore
+          zh_Hans: 亚洲/新加坡
+          pt_BR: Asia/Singapore
+      - value: Europe/London
+        label:
+          en_US: Europe/London
+          zh_Hans: 欧洲/伦敦
+          pt_BR: Europe/London
+      - value: Europe/Berlin
+        label:
+          en_US: Europe/Berlin
+          zh_Hans: 欧洲/柏林
+          pt_BR: Europe/Berlin
+      - value: Europe/Moscow
+        label:
+          en_US: Europe/Moscow
+          zh_Hans: 欧洲/莫斯科
+          pt_BR: Europe/Moscow
+      - value: Australia/Sydney
+        label:
+          en_US: Australia/Sydney
+          zh_Hans: 澳大利亚/悉尼
+          pt_BR: Australia/Sydney
+      - value: Pacific/Auckland
+        label:
+          en_US: Pacific/Auckland
+          zh_Hans: 太平洋/奥克兰
+          pt_BR: Pacific/Auckland
+      - value: Africa/Cairo
+        label:
+          en_US: Africa/Cairo
+          zh_Hans: 非洲/开罗
+          pt_BR: Africa/Cairo

dify/api/core/tools/builtin_tool/providers/time/tools/localtime_to_timestamp.py

@@ -0,0 +1,50 @@
+from collections.abc import Generator
+from datetime import datetime
+from typing import Any
+
+import pytz
+
+from core.tools.builtin_tool.tool import BuiltinTool
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.errors import ToolInvokeError
+
+
+class LocaltimeToTimestampTool(BuiltinTool):
+    def _invoke(
+        self,
+        user_id: str,
+        tool_parameters: dict[str, Any],
+        conversation_id: str | None = None,
+        app_id: str | None = None,
+        message_id: str | None = None,
+    ) -> Generator[ToolInvokeMessage, None, None]:
+        """
+        Convert localtime to timestamp
+        """
+        localtime = tool_parameters.get("localtime")
+        timezone = tool_parameters.get("timezone", "Asia/Shanghai")
+        if not timezone:
+            timezone = None
+        time_format = "%Y-%m-%d %H:%M:%S"
+
+        timestamp = self.localtime_to_timestamp(localtime, time_format, timezone)  # type: ignore
+        if not timestamp:
+            yield self.create_text_message(f"Invalid localtime: {localtime}")
+            return
+
+        yield self.create_text_message(f"{timestamp}")
+
+    # TODO: this method's type is messy
+    @staticmethod
+    def localtime_to_timestamp(localtime: str, time_format: str, local_tz=None) -> int | None:
+        try:
+            local_time = datetime.strptime(localtime, time_format)
+            if local_tz is None:
+                localtime = local_time.astimezone()  # type: ignore
+            elif isinstance(local_tz, str):
+                local_tz = pytz.timezone(local_tz)
+                localtime = local_tz.localize(local_time)  # type: ignore
+            timestamp = int(localtime.timestamp())  # type: ignore
+            return timestamp
+        except Exception as e:
+            raise ToolInvokeError(str(e))

dify/api/core/tools/builtin_tool/providers/time/tools/localtime_to_timestamp.yaml

@@ -0,0 +1,33 @@
+identity:
+  name: localtime_to_timestamp
+  author: zhuhao
+  label:
+    en_US: localtime to timestamp
+    zh_Hans: 获取时间戳
+description:
+  human:
+    en_US: A tool for localtime convert to timestamp
+    zh_Hans: 获取时间戳
+  llm: A tool for localtime convert to timestamp
+parameters:
+  - name: localtime
+    type: string
+    required: true
+    form: llm
+    label:
+      en_US: localtime
+      zh_Hans: 本地时间
+    human_description:
+      en_US: localtime, such as 2024-1-1 0:0:0
+      zh_Hans: 本地时间，比如 2024-1-1 0:0:0
+  - name: timezone
+    type: string
+    required: false
+    form: llm
+    label:
+      en_US: Timezone
+      zh_Hans: 时区
+    human_description:
+      en_US: Timezone, such as Asia/Shanghai
+      zh_Hans: 时区，比如 Asia/Shanghai
+    default: Asia/Shanghai

dify/api/core/tools/builtin_tool/providers/time/tools/timestamp_to_localtime.py

@@ -0,0 +1,49 @@
+from collections.abc import Generator
+from datetime import datetime
+from typing import Any
+
+import pytz
+
+from core.tools.builtin_tool.tool import BuiltinTool
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.errors import ToolInvokeError
+
+
+class TimestampToLocaltimeTool(BuiltinTool):
+    def _invoke(
+        self,
+        user_id: str,
+        tool_parameters: dict[str, Any],
+        conversation_id: str | None = None,
+        app_id: str | None = None,
+        message_id: str | None = None,
+    ) -> Generator[ToolInvokeMessage, None, None]:
+        """
+        Convert timestamp to localtime
+        """
+        timestamp: int = tool_parameters.get("timestamp", 0)
+        timezone = tool_parameters.get("timezone", "Asia/Shanghai")
+        if not timezone:
+            timezone = None
+        time_format = "%Y-%m-%d %H:%M:%S"
+
+        locatime = self.timestamp_to_localtime(timestamp, timezone)
+        if not locatime:
+            yield self.create_text_message(f"Invalid timestamp: {timestamp}")
+            return
+
+        localtime_format = locatime.strftime(time_format)
+
+        yield self.create_text_message(f"{localtime_format}")
+
+    @staticmethod
+    def timestamp_to_localtime(timestamp: int, local_tz=None) -> datetime | None:
+        try:
+            if local_tz is None:
+                local_tz = datetime.now().astimezone().tzinfo
+            if isinstance(local_tz, str):
+                local_tz = pytz.timezone(local_tz)
+            local_time = datetime.fromtimestamp(timestamp, local_tz)
+            return local_time
+        except Exception as e:
+            raise ToolInvokeError(str(e))

dify/api/core/tools/builtin_tool/providers/time/tools/timestamp_to_localtime.yaml

@@ -0,0 +1,33 @@
+identity:
+  name: timestamp_to_localtime
+  author: zhuhao
+  label:
+    en_US: Timestamp to localtime
+    zh_Hans: 时间戳转换
+description:
+  human:
+    en_US: A tool for timestamp convert to localtime
+    zh_Hans: 时间戳转换
+  llm: A tool for timestamp convert to localtime
+parameters:
+  - name: timestamp
+    type: number
+    required: true
+    form: llm
+    label:
+      en_US: Timestamp
+      zh_Hans: 时间戳
+    human_description:
+      en_US: Timestamp
+      zh_Hans: 时间戳
+  - name: timezone
+    type: string
+    required: false
+    form: llm
+    label:
+      en_US: Timezone
+      zh_Hans: 时区
+    human_description:
+      en_US: Timezone, such as Asia/Shanghai
+      zh_Hans: 时区，比如 Asia/Shanghai
+    default: Asia/Shanghai

dify/api/core/tools/builtin_tool/providers/time/tools/timezone_conversion.py

@@ -0,0 +1,53 @@
+from collections.abc import Generator
+from datetime import datetime
+from typing import Any
+
+import pytz
+
+from core.tools.builtin_tool.tool import BuiltinTool
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.errors import ToolInvokeError
+
+
+class TimezoneConversionTool(BuiltinTool):
+    def _invoke(
+        self,
+        user_id: str,
+        tool_parameters: dict[str, Any],
+        conversation_id: str | None = None,
+        app_id: str | None = None,
+        message_id: str | None = None,
+    ) -> Generator[ToolInvokeMessage, None, None]:
+        """
+        Convert time to equivalent time zone
+        """
+        current_time = tool_parameters.get("current_time")
+        current_timezone = tool_parameters.get("current_timezone", "Asia/Shanghai")
+        target_timezone = tool_parameters.get("target_timezone", "Asia/Tokyo")
+        target_time = self.timezone_convert(current_time, current_timezone, target_timezone)  # type: ignore
+        if not target_time:
+            yield self.create_text_message(
+                f"Invalid datetime and timezone: {current_time},{current_timezone},{target_timezone}"
+            )
+            return
+
+        yield self.create_text_message(f"{target_time}")
+
+    @staticmethod
+    def timezone_convert(current_time: str, source_timezone: str, target_timezone: str) -> str:
+        """
+        Convert a time string from source timezone to target timezone.
+        """
+        time_format = "%Y-%m-%d %H:%M:%S"
+        try:
+            # get source timezone
+            input_timezone = pytz.timezone(source_timezone)
+            # get target timezone
+            output_timezone = pytz.timezone(target_timezone)
+            local_time = datetime.strptime(current_time, time_format)
+            datetime_with_tz = input_timezone.localize(local_time)
+            # timezone convert
+            converted_datetime = datetime_with_tz.astimezone(output_timezone)
+            return converted_datetime.strftime(time_format)
+        except Exception as e:
+            raise ToolInvokeError(str(e))

dify/api/core/tools/builtin_tool/providers/time/tools/timezone_conversion.yaml

@@ -0,0 +1,44 @@
+identity:
+  name: timezone_conversion
+  author: zhuhao
+  label:
+    en_US: convert time to equivalent time zone
+    zh_Hans: 时区转换
+description:
+  human:
+    en_US: A tool to convert time to equivalent time zone
+    zh_Hans: 时区转换
+  llm: A tool to convert time to equivalent time zone
+parameters:
+  - name: current_time
+    type: string
+    required: true
+    form: llm
+    label:
+      en_US: current time
+      zh_Hans: 当前时间
+    human_description:
+      en_US: current time, such as 2024-1-1 0:0:0
+      zh_Hans: 当前时间，比如 2024-1-1 0:0:0
+  - name: current_timezone
+    type: string
+    required: true
+    form: llm
+    label:
+      en_US: Current Timezone
+      zh_Hans: 当前时区
+    human_description:
+      en_US: Current Timezone, such as Asia/Shanghai
+      zh_Hans: 当前时区，比如 Asia/Shanghai
+    default: Asia/Shanghai
+  - name: target_timezone
+    type: string
+    required: true
+    form: llm
+    label:
+      en_US: Target Timezone
+      zh_Hans: 目标时区
+    human_description:
+      en_US: Target Timezone, such as Asia/Tokyo
+      zh_Hans: 目标时区，比如 Asia/Tokyo
+    default: Asia/Tokyo

dify/api/core/tools/builtin_tool/providers/time/tools/weekday.py

@@ -0,0 +1,50 @@
+import calendar
+from collections.abc import Generator
+from datetime import datetime
+from typing import Any
+
+from core.tools.builtin_tool.tool import BuiltinTool
+from core.tools.entities.tool_entities import ToolInvokeMessage
+
+
+class WeekdayTool(BuiltinTool):
+    def _invoke(
+        self,
+        user_id: str,
+        tool_parameters: dict[str, Any],
+        conversation_id: str | None = None,
+        app_id: str | None = None,
+        message_id: str | None = None,
+    ) -> Generator[ToolInvokeMessage, None, None]:
+        """
+        Calculate the day of the week for a given date
+        """
+        year = tool_parameters.get("year")
+        month = tool_parameters.get("month")
+        if month is None:
+            raise ValueError("Month is required")
+        day = tool_parameters.get("day")
+
+        date_obj = self.convert_datetime(year, month, day)
+        if not date_obj:
+            yield self.create_text_message(f"Invalid date: Year {year}, Month {month}, Day {day}.")
+            return
+
+        weekday_name = calendar.day_name[date_obj.weekday()]
+        month_name = calendar.month_name[month]
+        readable_date = f"{month_name} {date_obj.day}, {date_obj.year}"
+        yield self.create_text_message(f"{readable_date} is {weekday_name}.")
+
+    @staticmethod
+    def convert_datetime(year, month, day) -> datetime | None:
+        try:
+            # allowed range in datetime module
+            if not (year >= 1 and 1 <= month <= 12 and 1 <= day <= 31):
+                return None
+
+            year = int(year)
+            month = int(month)
+            day = int(day)
+            return datetime(year, month, day)
+        except ValueError:
+            return None

dify/api/core/tools/builtin_tool/providers/time/tools/weekday.yaml

@@ -0,0 +1,42 @@
+identity:
+  name: weekday
+  author: Bowen Liang
+  label:
+    en_US: Weekday Calculator
+    zh_Hans: 星期几计算器
+description:
+  human:
+    en_US: A tool for calculating the weekday of a given date.
+    zh_Hans: 计算指定日期为星期几的工具。
+  llm: A tool for calculating the weekday of a given date by year, month and day.
+parameters:
+  - name: year
+    type: number
+    required: true
+    form: llm
+    label:
+      en_US: Year
+      zh_Hans: 年
+    human_description:
+      en_US: Year
+      zh_Hans: 年
+  - name: month
+    type: number
+    required: true
+    form: llm
+    label:
+      en_US: Month
+      zh_Hans: 月
+    human_description:
+      en_US: Month
+      zh_Hans: 月
+  - name: day
+    type: number
+    required: true
+    form: llm
+    label:
+      en_US: day
+      zh_Hans: 日
+    human_description:
+      en_US: day
+      zh_Hans: 日

dify/api/core/tools/builtin_tool/providers/webscraper/_assets/icon.svg

@@ -0,0 +1,3 @@
+<svg xmlns="http://www.w3.org/2000/svg" width="16" height="17" viewBox="0 0 16 17" fill="none">
+  <path fill-rule="evenodd" clip-rule="evenodd" d="M2.6665 1.16667C1.56193 1.16667 0.666504 2.0621 0.666504 3.16667C0.666504 4.27124 1.56193 5.16667 2.6665 5.16667C2.79161 5.16667 2.91403 5.15519 3.03277 5.13321C2.3808 6.09319 1.99984 7.25211 1.99984 8.5C1.99984 9.7479 2.3808 10.9068 3.03277 11.8668C2.91403 11.8448 2.79161 11.8333 2.6665 11.8333C1.56193 11.8333 0.666504 12.7288 0.666504 13.8333C0.666504 14.9379 1.56193 15.8333 2.6665 15.8333C3.77107 15.8333 4.6665 14.9379 4.6665 13.8333C4.6665 13.7082 4.65502 13.5858 4.63304 13.4671C5.59302 14.119 6.75194 14.5 7.99984 14.5C9.24773 14.5 10.4066 14.119 11.3666 13.4671C11.3447 13.5858 11.3332 13.7082 11.3332 13.8333C11.3332 14.9379 12.2286 15.8333 13.3332 15.8333C14.4377 15.8333 15.3332 14.9379 15.3332 13.8333C15.3332 12.7288 14.4377 11.8333 13.3332 11.8333C13.2081 11.8333 13.0856 11.8448 12.9669 11.8668C13.6189 10.9068 13.9998 9.7479 13.9998 8.5C13.9998 7.25211 13.6189 6.09319 12.9669 5.13321C13.0856 5.15519 13.2081 5.16667 13.3332 5.16667C14.4377 5.16667 15.3332 4.27124 15.3332 3.16667C15.3332 2.0621 14.4377 1.16667 13.3332 1.16667C12.2286 1.16667 11.3332 2.0621 11.3332 3.16667C11.3332 3.29177 11.3447 3.41419 11.3666 3.53293C10.4066 2.88097 9.24773 2.50001 7.99984 2.50001C6.75194 2.50001 5.59302 2.88097 4.63304 3.53293C4.65502 3.41419 4.6665 3.29177 4.6665 3.16667C4.6665 2.0621 3.77107 1.16667 2.6665 1.16667ZM3.38043 7.83334C3.63081 6.08287 4.85262 4.64578 6.48223 4.08565C5.79223 5.22099 5.36488 6.50185 5.23815 7.83334H3.38043ZM6.48228 12.9144C4.85264 12.3543 3.63082 10.9172 3.38043 9.16667H5.23815C5.3649 10.4982 5.79226 11.779 6.48228 12.9144ZM12.6192 9.16667C12.3689 10.9168 11.1475 12.3537 9.5183 12.9141C10.2082 11.7788 10.6355 10.498 10.7622 9.16667H12.6192ZM9.51834 4.08596C11.1475 4.64631 12.3689 6.0832 12.6192 7.83334H10.7622C10.6355 6.50197 10.2082 5.22123 9.51834 4.08596ZM9.4218 7.83334C9.27457 6.52262 8.78381 5.27411 8.00019 4.2145C7.21658 5.27411 6.72582 6.52262 6.57859 7.83334H9.4218ZM6.5786 9.16667C6.72583 10.4774 7.21659 11.7259 8.00019 12.7855C8.7838 11.7259 9.27456 10.4774 9.42179 9.16667H6.5786Z" fill="#DD2590"/>
+</svg>

dify/api/core/tools/builtin_tool/providers/webscraper/tools/webscraper.py

@@ -0,0 +1,39 @@
+from collections.abc import Generator
+from typing import Any
+
+from core.tools.builtin_tool.tool import BuiltinTool
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.errors import ToolInvokeError
+from core.tools.utils.web_reader_tool import get_url
+
+
+class WebscraperTool(BuiltinTool):
+    def _invoke(
+        self,
+        user_id: str,
+        tool_parameters: dict[str, Any],
+        conversation_id: str | None = None,
+        app_id: str | None = None,
+        message_id: str | None = None,
+    ) -> Generator[ToolInvokeMessage, None, None]:
+        """
+        invoke tools
+        """
+        try:
+            url = tool_parameters.get("url", "")
+            user_agent = tool_parameters.get("user_agent", "")
+            if not url:
+                yield self.create_text_message("Please input url")
+                return
+
+            # get webpage
+            result = get_url(url, user_agent=user_agent)
+
+            if tool_parameters.get("generate_summary"):
+                # summarize and return
+                yield self.create_text_message(self.summary(user_id=user_id, content=result))
+            else:
+                # return full webpage
+                yield self.create_text_message(result)
+        except Exception as e:
+            raise ToolInvokeError(str(e))

dify/api/core/tools/builtin_tool/providers/webscraper/tools/webscraper.yaml

@@ -0,0 +1,60 @@
+identity:
+  name: webscraper
+  author: Dify
+  label:
+    en_US: Web Scraper
+    zh_Hans: 网页爬虫
+    pt_BR: Web Scraper
+description:
+  human:
+    en_US: A tool for scraping webpages.
+    zh_Hans: 一个用于爬取网页的工具。
+    pt_BR: A tool for scraping webpages.
+  llm: A tool for scraping webpages. Input should be a URL.
+parameters:
+  - name: url
+    type: string
+    required: true
+    label:
+      en_US: URL
+      zh_Hans: 网页链接
+      pt_BR: URL
+    human_description:
+      en_US: used for linking to webpages
+      zh_Hans: 用于链接到网页
+      pt_BR: used for linking to webpages
+    llm_description: url for scraping
+    form: llm
+  - name: user_agent
+    type: string
+    required: false
+    label:
+      en_US: User Agent
+      zh_Hans: User Agent
+      pt_BR: User Agent
+    human_description:
+      en_US: used for identifying the browser.
+      zh_Hans: 用于识别浏览器。
+      pt_BR: used for identifying the browser.
+    form: form
+    default: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/100.0.1000.0 Safari/537.36
+  - name: generate_summary
+    type: boolean
+    required: false
+    label:
+      en_US: Whether to generate summary
+      zh_Hans: 是否生成摘要
+    human_description:
+      en_US: If true, the crawler will only return the page summary content.
+      zh_Hans: 如果启用，爬虫将仅返回页面摘要内容。
+    form: form
+    options:
+      - value: "true"
+        label:
+          en_US: "Yes"
+          zh_Hans: 是
+      - value: "false"
+        label:
+          en_US: "No"
+          zh_Hans: 否
+    default: "false"

dify/api/core/tools/builtin_tool/providers/webscraper/webscraper.py

@@ -0,0 +1,11 @@
+from typing import Any
+
+from core.tools.builtin_tool.provider import BuiltinToolProviderController
+
+
+class WebscraperProvider(BuiltinToolProviderController):
+    def _validate_credentials(self, user_id: str, credentials: dict[str, Any]):
+        """
+        Validate credentials
+        """
+        pass

dify/api/core/tools/builtin_tool/providers/webscraper/webscraper.yaml

@@ -0,0 +1,15 @@
+identity:
+  author: Dify
+  name: webscraper
+  label:
+    en_US: WebScraper
+    zh_Hans: 网页抓取
+    pt_BR: WebScraper
+  description:
+    en_US: Web Scrapper tool kit is used to scrape web
+    zh_Hans: 一个用于抓取网页的工具。
+    pt_BR: Web Scrapper tool kit is used to scrape web
+  icon: icon.svg
+  tags:
+    - productivity
+credentials_for_provider: {}

dify/api/core/tools/builtin_tool/tool.py

@@ -0,0 +1,148 @@
+from core.model_runtime.entities.llm_entities import LLMResult
+from core.model_runtime.entities.message_entities import PromptMessage, SystemPromptMessage, UserPromptMessage
+from core.tools.__base.tool import Tool
+from core.tools.__base.tool_runtime import ToolRuntime
+from core.tools.entities.tool_entities import ToolProviderType
+from core.tools.utils.model_invocation_utils import ModelInvocationUtils
+
+_SUMMARY_PROMPT = """You are a professional language researcher, you are interested in the language
+and you can quickly aimed at the main point of an webpage and reproduce it in your own words but
+retain the original meaning and keep the key points.
+however, the text you got is too long, what you got is possible a part of the text.
+Please summarize the text you got.
+"""
+
+
+class BuiltinTool(Tool):
+    """
+    Builtin tool
+
+    :param meta: the meta data of a tool call processing
+    """
+
+    def __init__(self, provider: str, **kwargs):
+        super().__init__(**kwargs)
+        self.provider = provider
+
+    def fork_tool_runtime(self, runtime: ToolRuntime) -> "BuiltinTool":
+        """
+        fork a new tool with metadata
+        :return: the new tool
+        """
+        return self.__class__(
+            entity=self.entity.model_copy(),
+            runtime=runtime,
+            provider=self.provider,
+        )
+
+    def invoke_model(self, user_id: str, prompt_messages: list[PromptMessage], stop: list[str]) -> LLMResult:
+        """
+        invoke model
+
+        :param user_id: the user id
+        :param prompt_messages: the prompt messages
+        :param stop: the stop words
+        :return: the model result
+        """
+        # invoke model
+        return ModelInvocationUtils.invoke(
+            user_id=user_id,
+            tenant_id=self.runtime.tenant_id or "",
+            tool_type="builtin",
+            tool_name=self.entity.identity.name,
+            prompt_messages=prompt_messages,
+        )
+
+    def tool_provider_type(self) -> ToolProviderType:
+        return ToolProviderType.BUILT_IN
+
+    def get_max_tokens(self) -> int:
+        """
+        get max tokens
+
+        :return: the max tokens
+        """
+        if self.runtime is None:
+            raise ValueError("runtime is required")
+
+        return ModelInvocationUtils.get_max_llm_context_tokens(
+            tenant_id=self.runtime.tenant_id or "",
+        )
+
+    def get_prompt_tokens(self, prompt_messages: list[PromptMessage]) -> int:
+        """
+        get prompt tokens
+
+        :param prompt_messages: the prompt messages
+        :return: the tokens
+        """
+        if self.runtime is None:
+            raise ValueError("runtime is required")
+
+        return ModelInvocationUtils.calculate_tokens(
+            tenant_id=self.runtime.tenant_id or "", prompt_messages=prompt_messages
+        )
+
+    def summary(self, user_id: str, content: str) -> str:
+        max_tokens = self.get_max_tokens()
+
+        if self.get_prompt_tokens(prompt_messages=[UserPromptMessage(content=content)]) < max_tokens * 0.6:
+            return content
+
+        def get_prompt_tokens(content: str) -> int:
+            return self.get_prompt_tokens(
+                prompt_messages=[SystemPromptMessage(content=_SUMMARY_PROMPT), UserPromptMessage(content=content)]
+            )
+
+        def summarize(content: str) -> str:
+            summary = self.invoke_model(
+                user_id=user_id,
+                prompt_messages=[SystemPromptMessage(content=_SUMMARY_PROMPT), UserPromptMessage(content=content)],
+                stop=[],
+            )
+
+            assert isinstance(summary.message.content, str)
+            return summary.message.content
+
+        lines = content.split("\n")
+        new_lines = []
+        # split long line into multiple lines
+        for i in range(len(lines)):
+            line = lines[i]
+            if not line.strip():
+                continue
+            if len(line) < max_tokens * 0.5:
+                new_lines.append(line)
+            elif get_prompt_tokens(line) > max_tokens * 0.7:
+                while get_prompt_tokens(line) > max_tokens * 0.7:
+                    new_lines.append(line[: int(max_tokens * 0.5)])
+                    line = line[int(max_tokens * 0.5) :]
+                new_lines.append(line)
+            else:
+                new_lines.append(line)
+
+        # merge lines into messages with max tokens
+        messages: list[str] = []
+        for j in new_lines:
+            if len(messages) == 0:
+                messages.append(j)
+            else:
+                if len(messages[-1]) + len(j) < max_tokens * 0.5:
+                    messages[-1] += j
+                if get_prompt_tokens(messages[-1] + j) > max_tokens * 0.7:
+                    messages.append(j)
+                else:
+                    messages[-1] += j
+
+        summaries = []
+        for i in range(len(messages)):
+            message = messages[i]
+            summary = summarize(message)
+            summaries.append(summary)
+
+        result = "\n".join(summaries)
+
+        if self.get_prompt_tokens(prompt_messages=[UserPromptMessage(content=result)]) > max_tokens * 0.7:
+            return self.summary(user_id=user_id, content=result)
+
+        return result