docs[minor]: LangGraph Migration Guide (#5487)

* [Docs] LangGraph Migration Guide * fixup * link * Update * Update and polish * Add to how to index page --------- Co-authored-by: jacoblee93 <jacoblee93@gmail.com>
langchain-ai · May 26, 2024 · 0a5988b · 0a5988b
1 parent ea22597
commit 0a5988b
Show file tree

Hide file tree

Showing 7 changed files with 1,715 additions and 64 deletions.
diff --git a/deno.json b/deno.json
@@ -3,7 +3,7 @@
     "langchain/": "npm:/langchain/",
     "@faker-js/faker": "npm:@faker-js/faker",
     "@langchain/anthropic": "npm:@langchain/anthropic",
-    "@langchain/community/": "npm:/@langchain/community/",
+    "@langchain/community/": "npm:/@langchain/community@0.2.2/",
     "@langchain/openai": "npm:@langchain/openai",
     "@langchain/cohere": "npm:@langchain/cohere",
     "@langchain/textsplitters": "npm:@langchain/textsplitters",
@@ -12,9 +12,11 @@
     "@langchain/core/": "npm:/@langchain/core/",
     "@langchain/pinecone": "npm:@langchain/pinecone",
     "@langchain/google-common": "npm:@langchain/google-common",
+    "@langchain/langgraph": "npm:/@langchain/langgraph@0.0.21",
+    "@langchain/langgraph/": "npm:/@langchain/langgraph@0.0.21/",
     "@microsoft/fetch-event-source": "npm:@microsoft/fetch-event-source",
     "@pinecone-database/pinecone": "npm:@pinecone-database/pinecone",
-    "cheerio": "npm:/cheerio",
+    "cheerio": "npm:cheerio",
     "chromadb": "npm:/chromadb",
     "dotenv/": "npm:/dotenv/",
     "zod": "npm:/zod",

diff --git a/docs/core_docs/.gitignore b/docs/core_docs/.gitignore
@@ -107,6 +107,8 @@ docs/how_to/output_parser_fixing.md
 docs/how_to/output_parser_fixing.mdx
 docs/how_to/multiple_queries.md
 docs/how_to/multiple_queries.mdx
+docs/how_to/migrate_agent.md
+docs/how_to/migrate_agent.mdx
 docs/how_to/logprobs.md
 docs/how_to/logprobs.mdx
 docs/how_to/lcel_cheatsheet.md

diff --git a/docs/core_docs/docs/how_to/agent_executor.ipynb b/docs/core_docs/docs/how_to/agent_executor.ipynb
@@ -100,7 +100,7 @@
     {
      "data": {
       "text/plain": [
-       "\u001b[32m`[{\"title\":\"Weather in San Francisco\",\"url\":\"https://www.weatherapi.com/\",\"content\":\"{'location': {'n`\u001b[39m... 1111 more characters"
+       "\u001b[32m`[{\"title\":\"Weather in San Francisco\",\"url\":\"https://www.weatherapi.com/\",\"content\":\"{'location': {'n`\u001b[39m... 1347 more characters"
       ]
      },
      "execution_count": 1,
@@ -109,6 +109,7 @@
     }
    ],
    "source": [
+    "import \"cheerio\"; // This is required in notebooks to use the `CheerioWebBaseLoader`\n",
     "import { TavilySearchResults } from \"@langchain/community/tools/tavily_search\"\n",
     "\n",
     "const search = new TavilySearchResults({\n",
@@ -152,24 +153,24 @@
     }
    ],
    "source": [
-    "import \"cheerio\"; // This is required in notebooks to use the `CheerioWebBaseLoader`\n",
-    "import { CheerioWebBaseLoader } from \"langchain/document_loaders/web/cheerio\";\n",
+    "import { CheerioWebBaseLoader } from \"@langchain/community/document_loaders/web/cheerio\";\n",
     "import { MemoryVectorStore } from \"langchain/vectorstores/memory\";\n",
     "import { OpenAIEmbeddings } from \"@langchain/openai\";\n",
     "import { RecursiveCharacterTextSplitter } from \"@langchain/textsplitters\";\n",
     "\n",
-    "const loader = new CheerioWebBaseLoader(\"https://docs.smith.langchain.com/overview\")\n",
-    "const docs = await loader.load()\n",
-    "const documents = await new RecursiveCharacterTextSplitter(\n",
-    "    {\n",
-    "        chunkSize: 1000,\n",
-    "        chunkOverlap: 200\n",
-    "    }\n",
-    ").splitDocuments(docs)\n",
-    "const vectorStore = await MemoryVectorStore.fromDocuments(documents, new OpenAIEmbeddings())\n",
+    "const loader = new CheerioWebBaseLoader(\"https://docs.smith.langchain.com/overview\");\n",
+    "const docs = await loader.load();\n",
+    "const splitter = new RecursiveCharacterTextSplitter(\n",
+    "  {\n",
+    "    chunkSize: 1000,\n",
+    "    chunkOverlap: 200\n",
+    "  }\n",
+    ");\n",
+    "const documents = await splitter.splitDocuments(docs);\n",
+    "const vectorStore = await MemoryVectorStore.fromDocuments(documents, new OpenAIEmbeddings());\n",
     "const retriever = vectorStore.asRetriever();\n",
     "\n",
-    "(await retriever.invoke(\"how to upload a dataset\"))[0]"
+    "(await retriever.invoke(\"how to upload a dataset\"))[0];"
    ]
   },
   {
@@ -258,6 +259,9 @@
     }
    ],
    "source": [
+    "import { ChatOpenAI } from \"@langchain/openai\";\n",
+    "const model = new ChatOpenAI({ model: \"gpt-4\", temperature: 0 })\n",
+    "\n",
     "import { HumanMessage } from \"@langchain/core/messages\";\n",
     "\n",
     "const response = await model.invoke([new HumanMessage(\"hi!\")]);\n",
@@ -336,9 +340,9 @@
       "  {\n",
       "    \"name\": \"tavily_search_results_json\",\n",
       "    \"args\": {\n",
-      "      \"input\": \"weather in San Francisco\"\n",
+      "      \"input\": \"current weather in San Francisco\"\n",
       "    },\n",
-      "    \"id\": \"call_y0nn6mbVCV5paX6RrqqFUqdC\"\n",
+      "    \"id\": \"call_VcSjZAZkEOx9lcHNZNXAjXkm\"\n",
       "  }\n",
       "]\n"
      ]
@@ -370,11 +374,7 @@
     "\n",
     "Now that we have defined the tools and the LLM, we can create the agent. We will be using a tool calling agent - for more information on this type of agent, as well as other options, see [this guide](/docs/concepts/#agent_types/).\n",
     "\n",
-    "We can first choose the prompt we want to use to guide the agent.\n",
-    "\n",
-    "If you want to see the contents of this prompt in the hub, you can go to:\n",
-    "\n",
-    "[https://smith.langchain.com/hub/hwchase17/openai-functions-agent](https://smith.langchain.com/hub/hwchase17/openai-functions-agent)"
+    "We can first choose the prompt we want to use to guide the agent:"
    ]
   },
   {
@@ -394,19 +394,18 @@
       "      prompt: PromptTemplate {\n",
       "        lc_serializable: true,\n",
       "        lc_kwargs: {\n",
-      "          template: \"You are a helpful assistant\",\n",
       "          inputVariables: [],\n",
       "          templateFormat: \"f-string\",\n",
-      "          partialVariables: {}\n",
+      "          template: \"You are a helpful assistant\"\n",
       "        },\n",
       "        lc_runnable: true,\n",
       "        name: undefined,\n",
       "        lc_namespace: [ \"langchain_core\", \"prompts\", \"prompt\" ],\n",
       "        inputVariables: [],\n",
       "        outputParser: undefined,\n",
-      "        partialVariables: {},\n",
-      "        template: \"You are a helpful assistant\",\n",
+      "        partialVariables: undefined,\n",
       "        templateFormat: \"f-string\",\n",
+      "        template: \"You are a helpful assistant\",\n",
       "        validateTemplate: true\n",
       "      }\n",
       "    },\n",
@@ -418,27 +417,26 @@
       "    prompt: PromptTemplate {\n",
       "      lc_serializable: true,\n",
       "      lc_kwargs: {\n",
-      "        template: \"You are a helpful assistant\",\n",
       "        inputVariables: [],\n",
       "        templateFormat: \"f-string\",\n",
-      "        partialVariables: {}\n",
+      "        template: \"You are a helpful assistant\"\n",
       "      },\n",
       "      lc_runnable: true,\n",
       "      name: undefined,\n",
       "      lc_namespace: [ \"langchain_core\", \"prompts\", \"prompt\" ],\n",
       "      inputVariables: [],\n",
       "      outputParser: undefined,\n",
-      "      partialVariables: {},\n",
-      "      template: \"You are a helpful assistant\",\n",
+      "      partialVariables: undefined,\n",
       "      templateFormat: \"f-string\",\n",
+      "      template: \"You are a helpful assistant\",\n",
       "      validateTemplate: true\n",
       "    },\n",
       "    messageClass: undefined,\n",
       "    chatMessageClass: undefined\n",
       "  },\n",
       "  MessagesPlaceholder {\n",
       "    lc_serializable: true,\n",
-      "    lc_kwargs: { optional: true, variableName: \"chat_history\" },\n",
+      "    lc_kwargs: { variableName: \"chat_history\", optional: true },\n",
       "    lc_runnable: true,\n",
       "    name: undefined,\n",
       "    lc_namespace: [ \"langchain_core\", \"prompts\", \"chat\" ],\n",
@@ -451,19 +449,18 @@
       "      prompt: PromptTemplate {\n",
       "        lc_serializable: true,\n",
       "        lc_kwargs: {\n",
-      "          template: \"{input}\",\n",
       "          inputVariables: [Array],\n",
       "          templateFormat: \"f-string\",\n",
-      "          partialVariables: {}\n",
+      "          template: \"{input}\"\n",
       "        },\n",
       "        lc_runnable: true,\n",
       "        name: undefined,\n",
       "        lc_namespace: [ \"langchain_core\", \"prompts\", \"prompt\" ],\n",
       "        inputVariables: [ \"input\" ],\n",
       "        outputParser: undefined,\n",
-      "        partialVariables: {},\n",
-      "        template: \"{input}\",\n",
+      "        partialVariables: undefined,\n",
       "        templateFormat: \"f-string\",\n",
+      "        template: \"{input}\",\n",
       "        validateTemplate: true\n",
       "      }\n",
       "    },\n",
@@ -475,43 +472,45 @@
       "    prompt: PromptTemplate {\n",
       "      lc_serializable: true,\n",
       "      lc_kwargs: {\n",
-      "        template: \"{input}\",\n",
       "        inputVariables: [ \"input\" ],\n",
       "        templateFormat: \"f-string\",\n",
-      "        partialVariables: {}\n",
+      "        template: \"{input}\"\n",
       "      },\n",
       "      lc_runnable: true,\n",
       "      name: undefined,\n",
       "      lc_namespace: [ \"langchain_core\", \"prompts\", \"prompt\" ],\n",
       "      inputVariables: [ \"input\" ],\n",
       "      outputParser: undefined,\n",
-      "      partialVariables: {},\n",
-      "      template: \"{input}\",\n",
+      "      partialVariables: undefined,\n",
       "      templateFormat: \"f-string\",\n",
+      "      template: \"{input}\",\n",
       "      validateTemplate: true\n",
       "    },\n",
       "    messageClass: undefined,\n",
       "    chatMessageClass: undefined\n",
       "  },\n",
       "  MessagesPlaceholder {\n",
       "    lc_serializable: true,\n",
-      "    lc_kwargs: { optional: false, variableName: \"agent_scratchpad\" },\n",
+      "    lc_kwargs: { variableName: \"agent_scratchpad\", optional: true },\n",
       "    lc_runnable: true,\n",
       "    name: undefined,\n",
       "    lc_namespace: [ \"langchain_core\", \"prompts\", \"chat\" ],\n",
       "    variableName: \"agent_scratchpad\",\n",
-      "    optional: false\n",
+      "    optional: true\n",
       "  }\n",
       "]\n"
      ]
     }
    ],
    "source": [
     "import { ChatPromptTemplate } from \"@langchain/core/prompts\";\n",
-    "import { pull } from \"langchain/hub\";\n",
     "\n",
-    "// Get the prompt to use - you can modify this!\n",
-    "const prompt = await pull<ChatPromptTemplate>(\"hwchase17/openai-functions-agent\");\n",
+    "const prompt = ChatPromptTemplate.fromMessages([\n",
+    "  [\"system\", \"You are a helpful assistant\"],\n",
+    "  [\"placeholder\", \"{chat_history}\"],\n",
+    "  [\"human\", \"{input}\"],\n",
+    "  [\"placeholder\", \"{agent_scratchpad}\"],\n",
+    "]);\n",
     "\n",
     "console.log(prompt.promptMessages);"
    ]
@@ -617,7 +616,9 @@
       "text/plain": [
        "{\n",
        "  input: \u001b[32m\"how can langsmith help with testing?\"\u001b[39m,\n",
-       "  output: \u001b[32m\"LangSmith can help with testing by providing a platform for building production-grade LLM applicatio\"\u001b[39m... 880 more characters\n",
+       "  output: \u001b[32m\"LangSmith can be a valuable tool for testing in several ways:\\n\"\u001b[39m +\n",
+       "    \u001b[32m\"\\n\"\u001b[39m +\n",
+       "    \u001b[32m\"1. **Logging Traces**: LangSmith prov\"\u001b[39m... 960 more characters\n",
        "}"
       ]
      },
@@ -651,7 +652,7 @@
       "text/plain": [
        "{\n",
        "  input: \u001b[32m\"whats the weather in sf?\"\u001b[39m,\n",
-       "  output: \u001b[32m\"The current weather in San Francisco is partly cloudy with a temperature of 64.0°F (17.8°C). The win\"\u001b[39m... 112 more characters\n",
+       "  output: \u001b[32m\"The current weather in San Francisco, California is partly cloudy with a temperature of 12.2°C (54.0\"\u001b[39m... 176 more characters\n",
        "}"
       ]
      },
@@ -753,7 +754,7 @@
        "    }\n",
        "  ],\n",
        "  input: \u001b[32m\"what's my name?\"\u001b[39m,\n",
-       "  output: \u001b[32m\"Your name is Bob! How can I help you, Bob?\"\u001b[39m\n",
+       "  output: \u001b[32m\"Your name is Bob. How can I assist you further?\"\u001b[39m\n",
        "}"
       ]
      },
@@ -785,8 +786,8 @@
     "\n",
     "Because we have multiple inputs, we need to specify two things:\n",
     "\n",
-    "- `input_messages_key`: The input key to use to add to the conversation history.\n",
-    "- `history_messages_key`: The key to add the loaded messages into.\n",
+    "- `inputMessagesKey`: The input key to use to add to the conversation history.\n",
+    "- `historyMessagesKey`: The key to add the loaded messages into.\n",
     "\n",
     "For more information on how to use this, see [this guide](/docs/how_to/message_history). "
    ]
@@ -819,22 +820,22 @@
        "    AIMessage {\n",
        "      lc_serializable: \u001b[33mtrue\u001b[39m,\n",
        "      lc_kwargs: {\n",
-       "        content: \u001b[32m\"Hello Bob! How can I assist you today?\"\u001b[39m,\n",
+       "        content: \u001b[32m\"Hello, Bob! How can I assist you today?\"\u001b[39m,\n",
        "        tool_calls: [],\n",
        "        invalid_tool_calls: [],\n",
        "        additional_kwargs: {},\n",
        "        response_metadata: {}\n",
        "      },\n",
        "      lc_namespace: [ \u001b[32m\"langchain_core\"\u001b[39m, \u001b[32m\"messages\"\u001b[39m ],\n",
-       "      content: \u001b[32m\"Hello Bob! How can I assist you today?\"\u001b[39m,\n",
+       "      content: \u001b[32m\"Hello, Bob! How can I assist you today?\"\u001b[39m,\n",
        "      name: \u001b[90mundefined\u001b[39m,\n",
        "      additional_kwargs: {},\n",
        "      response_metadata: {},\n",
        "      tool_calls: [],\n",
        "      invalid_tool_calls: []\n",
        "    }\n",
        "  ],\n",
-       "  output: \u001b[32m\"Hello Bob! How can I assist you today?\"\u001b[39m\n",
+       "  output: \u001b[32m\"Hello, Bob! How can I assist you today?\"\u001b[39m\n",
        "}"
       ]
      },
@@ -898,14 +899,14 @@
        "    AIMessage {\n",
        "      lc_serializable: \u001b[33mtrue\u001b[39m,\n",
        "      lc_kwargs: {\n",
-       "        content: \u001b[32m\"Hello Bob! How can I assist you today?\"\u001b[39m,\n",
+       "        content: \u001b[32m\"Hello, Bob! How can I assist you today?\"\u001b[39m,\n",
        "        tool_calls: [],\n",
        "        invalid_tool_calls: [],\n",
        "        additional_kwargs: {},\n",
        "        response_metadata: {}\n",
        "      },\n",
        "      lc_namespace: [ \u001b[32m\"langchain_core\"\u001b[39m, \u001b[32m\"messages\"\u001b[39m ],\n",
-       "      content: \u001b[32m\"Hello Bob! How can I assist you today?\"\u001b[39m,\n",
+       "      content: \u001b[32m\"Hello, Bob! How can I assist you today?\"\u001b[39m,\n",
        "      name: \u001b[90mundefined\u001b[39m,\n",
        "      additional_kwargs: {},\n",
        "      response_metadata: {},\n",
@@ -928,22 +929,22 @@
        "    AIMessage {\n",
        "      lc_serializable: \u001b[33mtrue\u001b[39m,\n",
        "      lc_kwargs: {\n",
-       "        content: \u001b[32m\"Your name is Bob! How can I help you, Bob?\"\u001b[39m,\n",
+       "        content: \u001b[32m\"Your name is Bob. How can I assist you further?\"\u001b[39m,\n",
        "        tool_calls: [],\n",
        "        invalid_tool_calls: [],\n",
        "        additional_kwargs: {},\n",
        "        response_metadata: {}\n",
        "      },\n",
        "      lc_namespace: [ \u001b[32m\"langchain_core\"\u001b[39m, \u001b[32m\"messages\"\u001b[39m ],\n",
-       "      content: \u001b[32m\"Your name is Bob! How can I help you, Bob?\"\u001b[39m,\n",
+       "      content: \u001b[32m\"Your name is Bob. How can I assist you further?\"\u001b[39m,\n",
        "      name: \u001b[90mundefined\u001b[39m,\n",
        "      additional_kwargs: {},\n",
        "      response_metadata: {},\n",
        "      tool_calls: [],\n",
        "      invalid_tool_calls: []\n",
        "    }\n",
        "  ],\n",
-       "  output: \u001b[32m\"Your name is Bob! How can I help you, Bob?\"\u001b[39m\n",
+       "  output: \u001b[32m\"Your name is Bob. How can I assist you further?\"\u001b[39m\n",
        "}"
       ]
      },
@@ -954,8 +955,8 @@
    ],
    "source": [
     "await agentWithChatHistory.invoke(\n",
-    "    { input: \"what's my name?\" },\n",
-    "    { configurable: { sessionId: \"<foo>\" }},\n",
+    "  { input: \"what's my name?\" },\n",
+    "  { configurable: { sessionId: \"<foo>\" }},\n",
     ")"
    ]
   },
@@ -972,12 +973,14 @@
    "id": "c029798f",
    "metadata": {},
    "source": [
-    "## Conclusion\n",
+    "## Next steps\n",
     "\n",
     "That's a wrap! In this quick start we covered how to create a simple agent. Agents are a complex topic, and there's lot to learn! \n",
     "\n",
     ":::{.callout-important}\n",
-    "This section covered building with LangChain Agents. LangChain Agents are fine for getting started, but past a certain point you will likely want flexibility and control that they do not offer. For working with more advanced agents, we'd recommend checking out [LangGraph](/docs/concepts/#langgraph)\n",
+    "This section covered building with LangChain Agents. LangChain Agents are fine for getting started, but past a certain point you will likely want flexibility and control that they do not offer. For working with more advanced agents, we'd recommend checking out [LangGraph](/docs/concepts/#langgraph).\n",
+    "\n",
+    "You can also see [this guide to help migrate to LangGraph](/docs/how_to/migrate_agent).\n",
     ":::"
    ]
   }

diff --git a/docs/core_docs/docs/how_to/index.mdx b/docs/core_docs/docs/how_to/index.mdx
@@ -163,6 +163,7 @@ For in depth how-to guides for agents, please check out [LangGraph](https://lang
 :::
 
 - [How to: use legacy LangChain Agents (AgentExecutor)](/docs/how_to/agent_executor)
+- [How to: migrate from legacy LangChain agents to LangGraph](/docs/how_to/migrate_agent)
 
 ### Callbacks