Squashed 'packages/tools/' content from commit 78317b9c

git-subtree-dir: packages/tools
git-subtree-split: 78317b9c127f18bd040c1d77e3c0840cdc9a5b38

crewai_tools/tools/rag/README.md

@@ -0,0 +1,61 @@
+# RagTool: A Dynamic Knowledge Base Tool
+
+RagTool is designed to answer questions by leveraging the power of RAG by leveraging (EmbedChain). It integrates seamlessly with the CrewAI ecosystem, offering a versatile and powerful solution for information retrieval.
+
+## **Overview**
+
+RagTool enables users to dynamically query a knowledge base, making it an ideal tool for applications requiring access to a vast array of information. Its flexible design allows for integration with various data sources, including files, directories, web pages, yoututbe videos and custom configurations.
+
+## **Usage**
+
+RagTool can be instantiated with data from different sources, including:
+
+- 📰 PDF file
+- 📊 CSV file
+- 📃 JSON file
+- 📝 Text
+- 📁 Directory/ Folder
+- 🌐 HTML Web page
+- 📽️ Youtube Channel
+- 📺 Youtube Video
+- 📚 Docs website
+- 📝 MDX file
+- 📄 DOCX file
+- 🧾 XML file
+- 📬 Gmail
+- 📝 Github
+- 🐘 Postgres
+- 🐬 MySQL
+- 🤖 Slack
+- 💬 Discord
+- 🗨️ Discourse
+- 📝 Substack
+- 🐝 Beehiiv
+- 💾 Dropbox
+- 🖼️ Image
+- ⚙️ Custom
+
+#### **Creating an Instance**
+
+```python
+from crewai_tools.tools.rag_tool import RagTool
+
+# Example: Loading from a file
+rag_tool = RagTool().from_file('path/to/your/file.txt')
+
+# Example: Loading from a directory
+rag_tool = RagTool().from_directory('path/to/your/directory')
+
+# Example: Loading from a web page
+rag_tool = RagTool().from_web_page('https://example.com')
+```
+
+## **Contribution**
+
+Contributions to RagTool and the broader CrewAI tools ecosystem are welcome. To contribute, please follow the standard GitHub workflow for forking the repository, making changes, and submitting a pull request.
+
+## **License**
+
+RagTool is open-source and available under the MIT license.
+
+Thank you for considering RagTool for your knowledge base needs. Your contributions and feedback are invaluable to making RagTool even better.

crewai_tools/tools/rag/rag_tool.py

@@ -0,0 +1,70 @@
+import portalocker
+
+from abc import ABC, abstractmethod
+from typing import Any
+from pydantic import BaseModel, ConfigDict, Field, model_validator
+
+from crewai.tools import BaseTool
+
+
+class Adapter(BaseModel, ABC):
+    model_config = ConfigDict(arbitrary_types_allowed=True)
+
+    @abstractmethod
+    def query(self, question: str) -> str:
+        """Query the knowledge base with a question and return the answer."""
+
+    @abstractmethod
+    def add(
+        self,
+        *args: Any,
+        **kwargs: Any,
+    ) -> None:
+        """Add content to the knowledge base."""
+
+
+class RagTool(BaseTool):
+    class _AdapterPlaceholder(Adapter):
+        def query(self, question: str) -> str:
+            raise NotImplementedError
+
+        def add(self, *args: Any, **kwargs: Any) -> None:
+            raise NotImplementedError
+
+    name: str = "Knowledge base"
+    description: str = "A knowledge base that can be used to answer questions."
+    summarize: bool = False
+    adapter: Adapter = Field(default_factory=_AdapterPlaceholder)
+    config: dict[str, Any] | None = None
+
+    @model_validator(mode="after")
+    def _set_default_adapter(self):
+        if isinstance(self.adapter, RagTool._AdapterPlaceholder):
+            try:
+                from embedchain import App
+            except ImportError:
+                raise ImportError("embedchain is not installed. Please install it with `pip install crewai-tools[embedchain]`")
+
+            from crewai_tools.adapters.embedchain_adapter import EmbedchainAdapter
+
+            with portalocker.Lock("crewai-rag-tool.lock", timeout=10):
+                app = App.from_config(config=self.config) if self.config else App()
+
+            self.adapter = EmbedchainAdapter(
+                embedchain_app=app, summarize=self.summarize
+            )
+
+        return self
+
+    def add(
+        self,
+        *args: Any,
+        **kwargs: Any,
+    ) -> None:
+        self.adapter.add(*args, **kwargs)
+
+    def _run(
+        self,
+        query: str,
+    ) -> str:
+        return f"Relevant Content:\n{self.adapter.query(query)}"