chore: import upstream snapshot with attribution

libs/cli/js-examples/src/agent/graph.ts

@@ -0,0 +1,104 @@
+/**
+ * Starter LangGraph.js Template
+ * Make this code your own!
+ */
+import { StateGraph } from "@langchain/langgraph";
+import { RunnableConfig } from "@langchain/core/runnables";
+import { StateAnnotation } from "./state.js";
+
+/**
+ * Define a node, these do the work of the graph and should have most of the logic.
+ * Must return a subset of the properties set in StateAnnotation.
+ * @param state The current state of the graph.
+ * @param config Extra parameters passed into the state graph.
+ * @returns Some subset of parameters of the graph state, used to update the state
+ * for the edges and nodes executed next.
+ */
+const callModel = async (
+  state: typeof StateAnnotation.State,
+  _config: RunnableConfig,
+): Promise<typeof StateAnnotation.Update> => {
+  /**
+   * Do some work... (e.g. call an LLM)
+   * For example, with LangChain you could do something like:
+   *
+   * ```bash
+   * $ npm i @langchain/anthropic
+   * ```
+   *
+   * ```ts
+   * import { ChatAnthropic } from "@langchain/anthropic";
+   * const model = new ChatAnthropic({
+   *   model: "claude-3-5-sonnet-20240620",
+   *   apiKey: process.env.ANTHROPIC_API_KEY,
+   * });
+   * const res = await model.invoke(state.messages);
+   * ```
+   *
+   * Or, with an SDK directly:
+   *
+   * ```bash
+   * $ npm i openai
+   * ```
+   *
+   * ```ts
+   * import OpenAI from "openai";
+   * const openai = new OpenAI({
+   *   apiKey: process.env.OPENAI_API_KEY,
+   * });
+   *
+   * const chatCompletion = await openai.chat.completions.create({
+   *   messages: [{
+   *     role: state.messages[0]._getType(),
+   *     content: state.messages[0].content,
+   *   }],
+   *   model: "gpt-4o-mini",
+   * });
+   * ```
+   */
+  console.log("Current state:", state);
+  return {
+    messages: [
+      {
+        role: "assistant",
+        content: `Hi there! How are you?`,
+      },
+    ],
+  };
+};
+
+/**
+ * Routing function: Determines whether to continue research or end the builder.
+ * This function decides if the gathered information is satisfactory or if more research is needed.
+ *
+ * @param state - The current state of the research builder
+ * @returns Either "callModel" to continue research or END to finish the builder
+ */
+export const route = (
+  state: typeof StateAnnotation.State,
+): "__end__" | "callModel" => {
+  if (state.messages.length > 0) {
+    return "__end__";
+  }
+  // Loop back
+  return "callModel";
+};
+
+// Finally, create the graph itself.
+const builder = new StateGraph(StateAnnotation)
+  // Add the nodes to do the work.
+  // Chaining the nodes together in this way
+  // updates the types of the StateGraph instance
+  // so you have static type checking when it comes time
+  // to add the edges.
+  .addNode("callModel", callModel)
+  // Regular edges mean "always transition to node B after node A is done"
+  // The "__start__" and "__end__" nodes are "virtual" nodes that are always present
+  // and represent the beginning and end of the builder.
+  .addEdge("__start__", "callModel")
+  // Conditional edges optionally route to different nodes (or end)
+  .addConditionalEdges("callModel", route);
+
+export const graph = builder.compile();
+
+graph.name = "New Agent";

libs/cli/js-examples/src/agent/state.ts

@@ -0,0 +1,59 @@
+import { BaseMessage, BaseMessageLike } from "@langchain/core/messages";
+import { Annotation, messagesStateReducer } from "@langchain/langgraph";
+
+/**
+ * A graph's StateAnnotation defines three main things:
+ * 1. The structure of the data to be passed between nodes (which "channels" to read from/write to and their types)
+ * 2. Default values for each field
+ * 3. Reducers for the state's. Reducers are functions that determine how to apply updates to the state.
+ * See [Reducers](https://langchain-ai.github.io/langgraphjs/concepts/low_level/#reducers) for more information.
+ */
+
+// This is the primary state of your agent, where you can store any information
+export const StateAnnotation = Annotation.Root({
+  /**
+   * Messages track the primary execution state of the agent.
+   *
+   * Typically accumulates a pattern of:
+   *
+   * 1. HumanMessage - user input
+   * 2. AIMessage with .tool_calls - agent picking tool(s) to use to collect
+   *     information
+   * 3. ToolMessage(s) - the responses (or errors) from the executed tools
+   *
+   *     (... repeat steps 2 and 3 as needed ...)
+   * 4. AIMessage without .tool_calls - agent responding in unstructured
+   *     format to the user.
+   *
+   * 5. HumanMessage - user responds with the next conversational turn.
+   *
+   *     (... repeat steps 2-5 as needed ... )
+   *
+   * Merges two lists of messages or message-like objects with role and content,
+   * updating existing messages by ID.
+   *
+   * Message-like objects are automatically coerced by `messagesStateReducer` into
+   * LangChain message classes. If a message does not have a given id,
+   * LangGraph will automatically assign one.
+   *
+   * By default, this ensures the state is "append-only", unless the
+   * new message has the same ID as an existing message.
+   *
+   * Returns:
+   *     A new list of messages with the messages from \`right\` merged into \`left\`.
+   *     If a message in \`right\` has the same ID as a message in \`left\`, the
+   *     message from \`right\` will replace the message from \`left\`.`
+   */
+  messages: Annotation<BaseMessage[], BaseMessageLike[]>({
+    reducer: messagesStateReducer,
+    default: () => [],
+  }),
+  /**
+   * Feel free to add additional attributes to your state as needed.
+   * Common examples include retrieved documents, extracted entities, API connections, etc.
+   *
+   * For simple fields whose value should be overwritten by the return value of a node,
+   * you don't need to define a reducer or default.
+   */
+  // additionalField: Annotation<string>,
+});