add prompt template tool and flow

update

update

update

docs/how-to-guides/develop-a-tool/add-prompt-template-for-tool.md

@@ -0,0 +1,51 @@
+# Add prompt template for tool
+Users sometimes need to use a prompt template within their tools. To simplify this, we've introduced the `PromptTemplate` feature.  
+In this guide, we will provide a detailed walkthrough on how to use `PromptTemplate` as a tool input. We will also demonstrate the user experience when utilizing this type of tool within a flow.
+
+## Prerequisites
+- Create a tool package as described in [Create and Use Tool Package](create-and-use-tool-package.md).
+- Ensure that the tool's type is `custom_llm`.
+
+## How to create a tool with prompt template
+Here we use [an existing tool package](../../../examples/tools/tool-package-quickstart/my_tool_package) as an example.  
+
+1. Add a `PromptTemplate` input for your tool, such as in [this example](../../../examples/tools/tool-package-quickstart/my_tool_package/tools/tool_with_prompt_template_input.py)
+
+    ```python
+    from jinja2 import Template
+    from promptflow import tool
+    from promptflow.connections import CustomConnection
+    # 1. import the PromptTemplate type
+    from promptflow.contracts.types import PromptTemplate
+
+
+    # 2. add a PromptTemplate input for your tool method
+    @tool
+    def my_tool(connection: CustomConnection, prompt: PromptTemplate, **kwargs) -> str:
+        # 3. customise your own code to handle and use the prompt here
+        message = Template(prompt, trim_blocks=True, keep_trailing_newline=True).render(**kwargs)
+        return message
+    ```
+
+2. Configure the tool YAML. Please note that the `PromptTemplate` input should not be included in the YAML. You can check out an example in [this location](../../../examples/tools/tool-package-quickstart/my_tool_package/yamls/tool_with_prompt_template_input.yaml):  
+
+    ```yaml
+    my_tool_package.tools.tool_with_prompt_template_input.my_tool:
+    name: Tool with PromptTemplate
+    description: This is a tool to demonstrate the usage of PromptTemplate
+    type: custom_llm
+    module: my_tool_package.tools.tool_with_prompt_template_input
+    function: my_tool
+    inputs:
+        connection:
+        type:
+            - CustomConnection
+    ```
+
+## Use a tool with a prompt template input in VS Code extension
+To use your tool with a prompt template input, follow the steps to [build and install your tool package](create-and-use-tool-package.md#build-and-share-the-tool-package) and [use your tool from VS Code extension](create-and-use-tool-package.md#use-your-tool-from-vscode-extension).  
+
+Here, we will use an existing flow to demonstrate the experience. Open [this flow](../../../examples/flows/standard/prompt-template-input-tool-showcase/flow.dag.yaml) in VS Code extension.  
+- There is a node named "tool_with_prompt_template" with a prompt template file, and the inputs of this node contain the input for the prompt template.  
+
+![use_prompt_template_in_flow](../../media/how-to-guides/develop-a-tool/use_prompt_template_in_flow.png)

docs/how-to-guides/develop-a-tool/create-and-use-tool-package.md

@@ -163,4 +163,5 @@ Alternatively, you can test your tool package using the script below to ensure t
 
 ## Advanced features
 [Customize your tool icon](add-a-tool-icon.md)  
-[Add category and tags for tool](add-category-and-tags-for-tool.md)  
+[Add category and tags for tool](add-category-and-tags-for-tool.md)  
+[Add Prompt Template for Custom LLM Tool](add-prompt-template-for-tool.md)  

docs/how-to-guides/develop-a-tool/index.md

@@ -8,4 +8,5 @@ We provide guides on how to develop a tool and use it.
 create-and-use-tool-package
 add-a-tool-icon
 add-category-and-tags-for-tool
+add-prompt-template-for-tool
 ```

examples/flows/standard/prompt-template-input-tool-showcase/flow.dag.yaml

@@ -0,0 +1,18 @@
+inputs:
+  text:
+    type: string
+    default: Microsoft
+outputs:
+  output:
+    type: string
+    reference: ${tool_with_prompt_template.output}
+nodes:
+- name: tool_with_prompt_template
+  type: custom_llm
+  source:
+    type: package_with_prompt
+    tool: my_tool_package.tools.tool_with_prompt_template_input.my_tool
+    path: prompt_template.jinja2
+  inputs:
+    connection: open_ai_connection
+    text: ${inputs.text}

examples/flows/standard/prompt-template-input-tool-showcase/prompt_template.jinja2

@@ -0,0 +1 @@
+Hello {{text}}.

examples/flows/standard/prompt-template-input-tool-showcase/requirements.txt

@@ -0,0 +1,3 @@
+promptflow
+promptflow-tools
+my-tools-package

examples/tools/tool-package-quickstart/my_tool_package/tools/tool_with_prompt_template_input.py

@@ -0,0 +1,14 @@
+from jinja2 import Template
+from promptflow import tool
+from promptflow.connections import CustomConnection
+from promptflow.contracts.types import PromptTemplate
+
+
+@tool
+def my_tool(connection: CustomConnection, prompt: PromptTemplate, **kwargs) -> str:
+    # Replace with your tool code, customise your own code to handle and use the prompt here.
+    # Usually connection contains configs to connect to an API.
+    # Use CustomConnection is a dict. You can use it like: connection.api_key, connection.api_base
+    # Not all tools need a connection. You can remove it if you don't need it.
+    message = Template(prompt, trim_blocks=True, keep_trailing_newline=True).render(**kwargs)
+    return message

examples/tools/tool-package-quickstart/my_tool_package/yamls/tool_with_prompt_template_input.yaml

@@ -0,0 +1,10 @@
+my_tool_package.tools.tool_with_prompt_template_input.my_tool:
+  name: Tool with PromptTemplate
+  description: This is a tool to demonstrate the usage of PromptTemplate
+  type: custom_llm
+  module: my_tool_package.tools.tool_with_prompt_template_input
+  function: my_tool
+  inputs:
+    connection:
+      type:
+        - CustomConnection

examples/tools/tool-package-quickstart/tests/test_tool_with_prompt_template_input.py

@@ -0,0 +1,28 @@
+import pytest
+import unittest
+
+from promptflow.connections import CustomConnection
+from my_tool_package.tools.tool_with_prompt_template_input import my_tool
+
+
+@pytest.fixture
+def my_custom_connection() -> CustomConnection:
+    my_custom_connection = CustomConnection(
+        {
+            "api-key" : "my-api-key",
+            "api-secret" : "my-api-secret",
+            "api-url" : "my-api-url"
+        }
+    )
+    return my_custom_connection
+
+
+class TestToolWithPromptTemplateInput:
+    def test_tool_with_prompt_template_input(self, my_custom_connection):
+        result = my_tool(my_custom_connection, "Hello {{text}}", text="Microsoft")
+        assert result == "Hello Microsoft"
+
+
+# Run the unit tests
+if __name__ == "__main__":
+    unittest.main()