Skip to content

Commit

Permalink
scripts[minor]docs[minor]: Add tools doc template & updated DuckDuckG…
Browse files Browse the repository at this point in the history 
…oSearch tool docs (#6386)

* scripts[minor]: Add tools doc template

* added duckduckgosearch tool

* chore: lint files

* cr

* cr
  • Loading branch information
@bracesproul
bracesproul authored Aug 6, 2024
1 parent fef0834 commit 0ed1646
Show file tree
Hide file tree
Showing 5 changed files with 793 additions and 47 deletions.
352 changes: 352 additions & 0 deletions docs/core_docs/docs/integrations/tools/duckduckgo_search.ipynb
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,352 @@
{
"cells": [
{
"cell_type": "raw",
"id": "10238e62-3465-4973-9279-606cbb7ccf16",
"metadata": {
"vscode": {
"languageId": "raw"
}
},
"source": [
"---\n",
"sidebar_label: DuckDuckGoSearch\n",
"---"
]
},
{
"cell_type": "markdown",
"id": "a6f91f20",
"metadata": {},
"source": [
"# DuckDuckGoSearch\n",
"\n",
"This notebook provides a quick overview for getting started with [DuckDuckGoSearch](/docs/integrations/tools/). For detailed documentation of all DuckDuckGoSearch features and configurations head to the [API reference](https://api.js.langchain.com/classes/langchain_community_tools_duckduckgo_search.DuckDuckGoSearch.html).\n",
"\n",
"DuckDuckGoSearch offers a privacy-focused search API designed for LLM Agents. It provides seamless integration with a wide range of data sources, prioritizing user privacy and relevant search results.\n",
"\n",
"## Overview\n",
"\n",
"### Integration details\n",
"\n",
"| Class | Package | Serializable | [PY support](https://python.langchain.com/docs/integrations/tools/ddg/) | Package latest |\n",
"| :--- | :--- | :---: | :---: | :---: |\n",
"| [DuckDuckGoSearch](https://api.js.langchain.com/classes/langchain_community_tools_duckduckgo_search.DuckDuckGoSearch.html) | [@langchain/community](https://api.js.langchain.com/modules/langchain_community_tools_duckduckgo_search.html) | beta | ✅ | ![NPM - Version](https://img.shields.io/npm/v/@langchain/community?style=flat-square&label=%20&) |\n",
"\n",
"## Setup\n",
"\n",
"The integration lives in the `@langchain/community` package, along with the `duck-duck-scrape` dependency:\n",
"\n",
"```{=mdx}\n",
"import IntegrationInstallTooltip from \"@mdx_components/integration_install_tooltip.mdx\";\n",
"import Npm2Yarn from \"@theme/Npm2Yarn\";\n",
"\n",
"<IntegrationInstallTooltip></IntegrationInstallTooltip>\n",
"\n",
"<Npm2Yarn>\n",
" @langchain/community duck-duck-scrape\n",
"</Npm2Yarn>\n",
"```\n",
"\n",
"### Credentials\n",
"\n",
"It's also helpful (but not needed) to set up [LangSmith](https://smith.langchain.com/) for best-in-class observability:\n",
"\n",
"```typescript\n",
"process.env.LANGCHAIN_TRACING_V2=\"true\"\n",
"process.env.LANGCHAIN_API_KEY=\"your-api-key\"\n",
"```"
]
},
{
"cell_type": "markdown",
"id": "1c97218f-f366-479d-8bf7-fe9f2f6df73f",
"metadata": {},
"source": [
"## Instantiation\n",
"\n",
"Here we show how to insatiate an instance of the `DuckDuckGoSearch` tool, with "
]
},
{
"cell_type": "code",
"execution_count": 1,
"id": "8b3ddfe9-ca79-494c-a7ab-1f56d9407a64",
"metadata": {},
"outputs": [],
"source": [
"import { DuckDuckGoSearch } from \"@langchain/community/tools/duckduckgo_search\"\n",
"\n",
"const tool = new DuckDuckGoSearch({ maxResults: 1 })"
]
},
{
"cell_type": "markdown",
"id": "74147a1a",
"metadata": {},
"source": [
"## Invocation\n",
"\n",
"### [Invoke directly with args](/docs/concepts/#invoke-with-just-the-arguments)"
]
},
{
"cell_type": "code",
"execution_count": 2,
"id": "65310a8b-eb0c-4d9e-a618-4f4abe2414fc",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"[{\"title\":\"Anthropic forecasts more than $850 mln in annualized revenue rate by ...\",\"link\":\"https://www.reuters.com/technology/anthropic-forecasts-more-than-850-mln-annualized-revenue-rate-by-2024-end-report-2023-12-26/\",\"snippet\":\"Dec 26 (Reuters) - Artificial intelligence startup <b>Anthropic</b> has projected it will generate more than $850 million in annualized <b>revenue</b> by the end of <b>2024</b>, the Information reported on Tuesday ...\"}]\n"
]
}
],
"source": [
"await tool.invoke(\"What is Anthropic's estimated revenue for 2024?\")"
]
},
{
"cell_type": "markdown",
"id": "d6e73897",
"metadata": {},
"source": [
"### [Invoke with ToolCall](/docs/concepts/#invoke-with-toolcall)\n",
"\n",
"We can also invoke the tool with a model-generated `ToolCall`, in which case a `ToolMessage` will be returned:"
]
},
{
"cell_type": "code",
"execution_count": 3,
"id": "f90e33a7",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"ToolMessage {\n",
" \"content\": \"[{\\\"title\\\":\\\"Anthropic forecasts more than $850 mln in annualized revenue rate by ...\\\",\\\"link\\\":\\\"https://www.reuters.com/technology/anthropic-forecasts-more-than-850-mln-annualized-revenue-rate-by-2024-end-report-2023-12-26/\\\",\\\"snippet\\\":\\\"Dec 26 (Reuters) - Artificial intelligence startup <b>Anthropic</b> has projected it will generate more than $850 million in annualized <b>revenue</b> by the end of <b>2024</b>, the Information reported on Tuesday ...\\\"}]\",\n",
" \"name\": \"duckduckgo-search\",\n",
" \"additional_kwargs\": {},\n",
" \"response_metadata\": {},\n",
" \"tool_call_id\": \"tool_call_id\"\n",
"}\n"
]
}
],
"source": [
"// This is usually generated by a model, but we'll create a tool call directly for demo purposes.\n",
"const modelGeneratedToolCall = {\n",
" args: {\n",
" input: \"What is Anthropic's estimated revenue for 2024?\"\n",
" },\n",
" id: \"tool_call_id\",\n",
" name: tool.name,\n",
" type: \"tool_call\",\n",
"}\n",
"await tool.invoke(modelGeneratedToolCall)"
]
},
{
"cell_type": "markdown",
"id": "659f9fbd-6fcf-445f-aa8c-72d8e60154bd",
"metadata": {},
"source": [
"## Chaining\n",
"\n",
"We can use our tool in a chain by first binding it to a [tool-calling model](/docs/how_to/tool_calling/) and then calling it:\n",
"\n",
"```{=mdx}\n",
"import ChatModelTabs from \"@theme/ChatModelTabs\";\n",
"\n",
"<ChatModelTabs customVarName=\"llm\" />\n",
"```\n"
]
},
{
"cell_type": "code",
"execution_count": 4,
"id": "af3123ad-7a02-40e5-b58e-7d56e23e5830",
"metadata": {},
"outputs": [],
"source": [
"// @lc-docs-hide-cell\n",
"\n",
"import { ChatOpenAI } from \"@langchain/openai\"\n",
"\n",
"const llm = new ChatOpenAI({\n",
" model: \"gpt-4o-mini\",\n",
"})"
]
},
{
"cell_type": "code",
"execution_count": 12,
"id": "fdbf35b5-3aaf-4947-9ec6-48c21533fb95",
"metadata": {},
"outputs": [],
"source": [
"import { ChatPromptTemplate } from \"@langchain/core/prompts\"\n",
"import { RunnableConfig } from \"@langchain/core/runnables\"\n",
"import { AIMessage } from \"@langchain/core/messages\"\n",
"\n",
"const prompt = ChatPromptTemplate.fromMessages(\n",
" [\n",
" [\"system\", \"You are a helpful assistant.\"],\n",
" [\"human\", \"{user_input}\"],\n",
" [\"placeholder\", \"{messages}\"],\n",
" ]\n",
")\n",
"\n",
"// specifying tool_choice will force the model to call this tool.\n",
"const llmWithTools = llm.bindTools([tool], {\n",
" tool_choice: tool.name\n",
"})\n",
"\n",
"const llmChain = prompt.pipe(llmWithTools);\n",
"\n",
"const toolChain = async (userInput: string, config?: RunnableConfig): Promise<AIMessage> => {\n",
" const input_ = { user_input: userInput };\n",
" const aiMsg = await llmChain.invoke(input_, config);\n",
" const toolMsgs = await tool.batch(aiMsg.tool_calls, config);\n",
" return llmChain.invoke({ ...input_, messages: [aiMsg, ...toolMsgs] }, config);\n",
"};\n",
"\n",
"const toolChainResult = await toolChain(\"What is Anthropic's estimated revenue for 2024?\");"
]
},
{
"cell_type": "code",
"execution_count": 13,
"id": "28448fe2",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"AIMessage {\n",
" \"tool_calls\": [\n",
" {\n",
" \"name\": \"duckduckgo-search\",\n",
" \"args\": {\n",
" \"input\": \"Anthropic revenue 2024 forecast\"\n",
" },\n",
" \"type\": \"tool_call\",\n",
" \"id\": \"call_E22L1T1bI6xPrMtL8wrKW5C5\"\n",
" }\n",
" ],\n",
" \"content\": \"\"\n",
"}\n"
]
}
],
"source": [
"const { tool_calls, content } = toolChainResult;\n",
"\n",
"console.log(\"AIMessage\", JSON.stringify({\n",
" tool_calls,\n",
" content\n",
"}, null, 2))"
]
},
{
"cell_type": "markdown",
"id": "71ceafa7",
"metadata": {},
"source": [
"## With an agent\n",
"\n",
"We can also pass the `DuckDuckGoSearch` tool to an agent:"
]
},
{
"cell_type": "code",
"execution_count": 14,
"id": "77a05a61",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"{\n",
" input: \"What is Anthropic's estimated revenue for 2024?\",\n",
" output: 'Anthropic has projected that it will generate more than $850 million in annualized revenue by the end of 2024.'\n",
"}\n"
]
}
],
"source": [
"import { DuckDuckGoSearch } from \"@langchain/community/tools/duckduckgo_search\";\n",
"import { ChatOpenAI } from \"@langchain/openai\";\n",
"import type { ChatPromptTemplate } from \"@langchain/core/prompts\";\n",
"\n",
"import { pull } from \"langchain/hub\";\n",
"import { AgentExecutor, createOpenAIFunctionsAgent } from \"langchain/agents\";\n",
"\n",
"// Define the tools the agent will have access to.\n",
"const toolsForAgent = [new DuckDuckGoSearch({ maxResults: 1 })];\n",
"\n",
"// Get the prompt to use - you can modify this!\n",
"// If you want to see the prompt in full, you can at:\n",
"// https://smith.langchain.com/hub/hwchase17/openai-functions-agent\n",
"const promptForAgent = await pull<ChatPromptTemplate>(\n",
" \"hwchase17/openai-functions-agent\"\n",
");\n",
"const llmForAgent = new ChatOpenAI({\n",
" model: \"gpt-4-turbo-preview\",\n",
" temperature: 0,\n",
"});\n",
"const agent = await createOpenAIFunctionsAgent({\n",
" llm: llmForAgent,\n",
" tools: toolsForAgent,\n",
" prompt: promptForAgent,\n",
"});\n",
"const agentExecutor = new AgentExecutor({\n",
" agent,\n",
" tools: toolsForAgent,\n",
"});\n",
"const agentResult = await agentExecutor.invoke({\n",
" input: \"What is Anthropic's estimated revenue for 2024?\",\n",
"});\n",
"\n",
"console.log(agentResult);"
]
},
{
"cell_type": "markdown",
"id": "4ac8146c",
"metadata": {},
"source": [
"## API reference\n",
"\n",
"For detailed documentation of all DuckDuckGoSearch features and configurations head to the [API reference](https://api.js.langchain.com/classes/langchain_community_tools_duckduckgo_search.DuckDuckGoSearch.html)"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "TypeScript",
"language": "typescript",
"name": "tslab"
},
"language_info": {
"codemirror_mode": {
"mode": "typescript",
"name": "javascript",
"typescript": true
},
"file_extension": ".ts",
"mimetype": "text/typescript",
"name": "typescript",
"version": "3.7.2"
}
},
"nbformat": 4,
"nbformat_minor": 5
}
Loading

0 comments on commit 0ed1646

Please sign in to comment.