@axiom-lattice/core 1.0.1

package/src/deep_agent/graph.ts

@@ -0,0 +1,106 @@
+/**
+ * Main createDeepAgent function for Deep Agents
+ *
+ * Main entry point for creating deep agents with TypeScript types for all parameters:
+ * tools, instructions, model, subagents, and stateSchema. Combines built-in tools with
+ * provided tools, creates task tool using createTaskTool(), and returns createReactAgent
+ * with proper configuration. Ensures exact parameter matching and behavior with Python version.
+ */
+
+// import "@langchain/anthropic/zod";
+import { createTaskTool } from "./subAgent";
+import { writeTodos, readFile, writeFile, editFile, ls } from "./tools";
+import type { CreateDeepAgentParams } from "./types";
+import type { StructuredTool } from "@langchain/core/tools";
+import { z } from "zod";
+import { DeepAgentState } from "./state";
+import { getModelLattice } from "../model_lattice/ModelLatticeManager";
+import { createReactAgent } from "@langchain/langgraph/prebuilt";
+import { getCheckpointSaver } from "@memory_lattice/MemoryLatticeManager";
+
+/**
+ * Base prompt that provides instructions about available tools
+ * Ported from Python implementation to ensure consistent behavior
+ */
+const BASE_PROMPT = `You have access to a number of standard tools
+
+## \`write_todos\`
+
+You have access to the \`write_todos\` tools to help you manage and plan tasks. Use these tools VERY frequently to ensure that you are tracking your tasks and giving the user visibility into your progress.
+These tools are also EXTREMELY helpful for planning tasks, and for breaking down larger complex tasks into smaller steps. If you do not use this tool when planning, you may forget to do important tasks - and that is unacceptable.
+
+It is critical that you mark todos as completed as soon as you are done with a task. Do not batch up multiple tasks before marking them as completed.
+## \`task\`
+
+- When doing web search, prefer to use the \`task\` tool in order to reduce context usage.`;
+
+/**
+ * Built-in tools that are always available in Deep Agents
+ */
+const BUILTIN_TOOLS: StructuredTool[] = [
+  writeTodos,
+  readFile,
+  writeFile,
+  editFile,
+  ls,
+];
+
+/**
+ * Create a Deep Agent with TypeScript types for all parameters.
+ * Combines built-in tools with provided tools, creates task tool using createTaskTool(),
+ * and returns createReactAgent with proper configuration.
+ * Ensures exact parameter matching and behavior with Python version.
+ */
+export function createDeepAgent<
+  StateSchema extends z.ZodObject<any, any, any, any, any>
+>(params: CreateDeepAgentParams<StateSchema> = {}) {
+  const {
+    tools = [],
+    instructions,
+    model = getModelLattice("default")?.client,
+    subagents = [],
+  } = params;
+
+  if (!model) {
+    throw new Error("Model not found");
+  }
+
+  const stateSchema = params.stateSchema
+    ? DeepAgentState.extend(params.stateSchema.shape)
+    : DeepAgentState;
+
+  // Combine built-in tools with provided tools
+  const allTools: StructuredTool[] = [...BUILTIN_TOOLS, ...tools];
+  // Create task tool using createTaskTool() if subagents are provided
+  if (subagents.length > 0) {
+    // Create tools map for task tool creation
+    const toolsMap: Record<string, StructuredTool> = {};
+    for (const tool of allTools) {
+      if (tool.name) {
+        toolsMap[tool.name] = tool;
+      }
+    }
+
+    const taskTool = createTaskTool({
+      subagents,
+      tools: toolsMap,
+      model,
+      stateSchema,
+    });
+    allTools.push(taskTool);
+  }
+
+  // Combine instructions with base prompt like Python implementation
+  const finalInstructions = instructions
+    ? instructions + BASE_PROMPT
+    : BASE_PROMPT;
+
+  // Return createReactAgent with proper configuration
+  return createReactAgent({
+    llm: model,
+    tools: allTools,
+    stateSchema: stateSchema,
+    prompt: finalInstructions,
+    checkpointer: getCheckpointSaver("default"),
+  });
+}

package/src/deep_agent/index.ts

@@ -0,0 +1,25 @@
+/**
+ * Deep Agents TypeScript Implementation
+ *
+ * A TypeScript port of the Python Deep Agents library for building controllable AI agents with LangGraph.
+ * This implementation maintains 1:1 compatibility with the Python version.
+ */
+
+export { createDeepAgent } from "./graph";
+export { writeTodos, readFile, writeFile, editFile, ls } from "./tools";
+export { DeepAgentState, fileReducer } from "./state";
+export {
+  WRITE_TODOS_DESCRIPTION,
+  TASK_DESCRIPTION_PREFIX,
+  TASK_DESCRIPTION_SUFFIX,
+  EDIT_DESCRIPTION,
+  TOOL_DESCRIPTION,
+} from "./prompts";
+export type {
+  SubAgent,
+  Todo,
+  DeepAgentStateType,
+  CreateDeepAgentParams,
+  CreateTaskToolParams,
+  TodoStatus,
+} from "./types";

package/src/deep_agent/prompts.ts

@@ -0,0 +1,284 @@
+/**
+ * Prompt constants for Deep Agents
+ *
+ * All prompt strings ported from Python implementation with exact string content and formatting
+ * to ensure 1:1 compatibility with the Python version.
+ */
+
+/**
+ * Description for the write_todos tool
+ * Ported exactly from Python WRITE_TODOS_DESCRIPTION
+ */
+export const WRITE_TODOS_DESCRIPTION = `Use this tool to create and manage a structured task list for your current work session. This helps you track progress, organize complex tasks, and demonstrate thoroughness to the user. It also helps the user understand the progress of the task and overall progress of their requests.
+
+When to Use This Tool
+Use this tool proactively in these scenarios:
+
+Complex multi-step tasks - When a task requires 3 or more distinct steps or actions
+Non-trivial and complex tasks - Tasks that require careful planning or multiple operations
+User explicitly requests todo list - When the user directly asks you to use the todo list
+User provides multiple tasks - When users provide a list of things to be done (numbered or comma-separated)
+After receiving new instructions - Immediately capture user requirements as todos
+When you start working on a task - Mark it as in_progress BEFORE beginning work. Ideally you should only have one todo as in_progress at a time
+After completing a task - Mark it as completed and add any new follow-up tasks discovered during implementation
+When NOT to Use This Tool
+Skip using this tool when:
+
+There is only a single, straightforward task
+The task is trivial and tracking it provides no organizational benefit
+The task can be completed in less than 3 trivial steps
+The task is purely conversational or informational
+NOTE that you should not use this tool if there is only one trivial task to do. In this case you are better off just doing the task directly.
+
+Examples of When to Use the Todo List
+<example>
+User: I want to add a dark mode toggle to the application settings. Make sure you run the tests and build when you're done!
+Assistant: I'll help add a dark mode toggle to your application settings. Let me create a todo list to track this implementation.
+*Creates todo list with the following items:*
+1. Create dark mode toggle component in Settings page
+2. Add dark mode state management (context/store)
+3. Implement CSS-in-JS styles for dark theme
+4. Update existing components to support theme switching
+5. Run tests and build process, addressing any failures or errors that occur
+*Begins working on the first task*
+<reasoning>
+The assistant used the todo list because:
+1. Adding dark mode is a multi-step feature requiring UI, state management, and styling changes
+2. The user explicitly requested tests and build be run afterward
+3. The assistant inferred that tests and build need to pass by adding "Ensure tests and build succeed" as the final task
+</reasoning>
+</example>
+<example>
+User: Help me plan a comprehensive marketing campaign for our new product launch, including social media, email marketing, and press releases.
+Assistant: I'll help you plan a comprehensive marketing campaign for your product launch. Let me create a todo list to organize all the components.
+*Creates todo list with the following items:*
+1. Define target audience and key messaging
+2. Create social media content calendar and posting schedule
+3. Design email marketing sequence and automation flows
+4. Draft press release and identify media contacts
+5. Set up tracking and analytics for campaign performance
+6. Create visual assets and brand materials
+7. Coordinate launch timeline and deliverables
+*Begins working on the first task*
+<reasoning>
+The assistant used the todo list because:
+1. Marketing campaign planning involves multiple distinct channels and activities
+2. Each component requires careful coordination and planning
+3. The systematic approach ensures all aspects of the launch are covered
+4. Progress tracking helps maintain timeline and deliverables
+</reasoning>
+</example>
+<example>
+User: Help me rename the function getCwd to getCurrentWorkingDirectory across my project
+Assistant: Let me first search through your codebase to find all occurrences of 'getCwd'.
+*Uses search tools to locate all instances of getCwd in the codebase*
+Assistant: I've found 15 instances of 'getCwd' across 8 different files. Let me create a todo list to track these changes.
+*Creates todo list with specific items for each file that needs updating*
+<reasoning>
+The assistant used the todo list because:
+1. First, the assistant searched to understand the scope of the task
+2. Upon finding multiple occurrences across different files, it determined this was a complex task with multiple steps
+3. The todo list helps ensure every instance is tracked and updated systematically
+4. This approach prevents missing any occurrences and maintains consistency
+</reasoning>
+</example>
+<example>
+User: I need help organizing my home office: decluttering the desk, setting up a filing system, organizing digital files, and creating a daily workflow.
+Assistant: I'll help you organize your home office systematically. Let me create a todo list to track each area of organization.
+*Creates todo list with items like: 1) Sort and declutter desk items, 2) Set up physical filing system for documents, 3) Organize digital files and folders, 4) Create daily workflow and productivity systems, 5) Establish maintenance routines*
+Let's start with decluttering your desk space to create a clean foundation for the other organizational systems.
+<reasoning>
+The assistant used the todo list because:
+1. The user provided multiple distinct organizational tasks
+2. Each area requires different approaches and careful planning
+3. Home organization benefits from systematic, step-by-step completion
+4. Progress tracking ensures all areas are addressed thoroughly
+</reasoning>
+</example>
+Examples of When NOT to Use the Todo List
+<example>
+User: How do I print 'Hello World' in Python?
+Assistant: In Python, you can print "Hello World" with this simple code:
+
+\`\`\`python
+print("Hello World")
+\`\`\`
+
+This will output the text "Hello World" to the console when executed.
+<reasoning>
+The assistant did not use the todo list because this is a single, trivial task that can be completed in one step. There's no need to track multiple tasks or steps for such a straightforward request.
+</reasoning>
+</example>
+<example>
+User: What's the best way to water houseplants?
+Assistant: The best way to water houseplants depends on the specific plant type, but here are some general guidelines:
+
+Check soil moisture by inserting your finger about 1-2 inches into the soil
+Water thoroughly when the top inch feels dry, allowing water to drain from the bottom
+Use room temperature water and water in the morning when possible
+Avoid overwatering, which is more harmful than underwatering for most plants
+
+Different plants have different needs, so it's important to research your specific varieties for optimal care.
+
+<reasoning>
+The assistant did not use the todo list because this is an informational request with no actual tasks to complete. The user is simply asking for advice and explanation, not for the assistant to perform multiple steps or activities.
+</reasoning>
+</example>
+<example>
+User: Can you add a comment to the calculateTotal function to explain what it does?
+Assistant: Sure, let me add a comment to the calculateTotal function to explain what it does.
+* Uses the Edit tool to add a comment to the calculateTotal function *
+<reasoning>
+The assistant did not use the todo list because this is a single, straightforward task confined to one location. Adding a comment doesn't require tracking multiple steps or systematic organization.
+</reasoning>
+</example>
+<example>
+User: What time is it in Tokyo right now?
+Assistant: I'll check the current time in Tokyo for you.
+
+*Searches for current time in Tokyo*
+
+The current time in Tokyo, Japan is [current time]. Tokyo is in the Japan Standard Time (JST) zone, which is UTC+9.
+
+<reasoning>
+The assistant did not use the todo list because this is a single information lookup with immediate results. There are no multiple steps to track or organize, making the todo list unnecessary for this straightforward request.
+</reasoning>
+</example>
+Task States and Management
+Task States: Use these states to track progress:
+
+pending: Task not yet started
+in_progress: Currently working on (limit to ONE task at a time)
+completed: Task finished successfully
+Task Management:
+
+Update task status in real-time as you work
+Mark tasks complete IMMEDIATELY after finishing (don't batch completions)
+Only have ONE task in_progress at any time
+Complete current tasks before starting new ones
+Remove tasks that are no longer relevant from the list entirely
+Task Completion Requirements:
+
+ONLY mark a task as completed when you have FULLY accomplished it
+If you encounter errors, blockers, or cannot finish, keep the task as in_progress
+When blocked, create a new task describing what needs to be resolved
+Never mark a task as completed if:
+There are unresolved issues or errors
+Work is partial or incomplete
+You encountered blockers that prevent completion
+You couldn't find necessary resources or dependencies
+Quality standards haven't been met
+Task Breakdown:
+
+Create specific, actionable items
+Break complex tasks into smaller, manageable steps
+Use clear, descriptive task names
+When in doubt, use this tool. Being proactive with task management demonstrates attentiveness and ensures you complete all requirements successfully.`;
+
+/**
+ * Prefix for task tool description
+ * Ported exactly from Python TASK_DESCRIPTION_PREFIX
+ */
+export const TASK_DESCRIPTION_PREFIX = `Launch a new agent to handle complex, multi-step tasks autonomously.
+
+Available agent types and the tools they have access to:
+
+general-purpose: General-purpose agent for researching complex questions, searching for files and content, and executing multi-step tasks. When you are searching for a keyword or file and are not confident that you will find the right match in the first few tries use this agent to perform the search for you. (Tools: *)
+{other_agents}
+`;
+
+/**
+ * Suffix for task tool description
+ * Ported exactly from Python TASK_DESCRIPTION_SUFFIX
+ */
+export const TASK_DESCRIPTION_SUFFIX = `When using the Task tool, you must specify a subagent_type parameter to select which agent type to use.
+
+When to use the Agent tool:
+
+When you are instructed to execute custom slash commands. Use the Agent tool with the slash command invocation as the entire prompt. The slash command can take arguments. For example: Task(description="Check the file", prompt="/check-file path/to/file.py")
+When NOT to use the Agent tool:
+
+If you want to read a specific file path, use the Read or Glob tool instead of the Agent tool, to find the match more quickly
+If you are searching for a specific term or definition within a known location, use the Glob tool instead, to find the match more quickly
+If you are searching for content within a specific file or set of 2-3 files, use the Read tool instead of the Agent tool, to find the match more quickly
+Other tasks that are not related to the agent descriptions above
+Usage notes:
+
+Launch multiple agents concurrently whenever possible, to maximize performance; to do that, use a single message with multiple tool uses
+When the agent is done, it will return a single message back to you. The result returned by the agent is not visible to the user. To show the user the result, you should send a text message back to the user with a concise summary of the result.
+Each agent invocation is stateless. You will not be able to send additional messages to the agent, nor will the agent be able to communicate with you outside of its final report. Therefore, your prompt should contain a highly detailed task description for the agent to perform autonomously and you should specify exactly what information the agent should return back to you in its final and only message to you.
+The agent's outputs should generally be trusted
+Clearly tell the agent whether you expect it to create content, perform analysis, or just do research (search, file reads, web fetches, etc.), since it is not aware of the user's intent
+If the agent description mentions that it should be used proactively, then you should try your best to use it without the user having to ask for it first. Use your judgement.
+Example usage:
+
+<example_agent_descriptions>
+"content-reviewer": use this agent after you are done creating significant content or documents
+"greeting-responder": use this agent when to respond to user greetings with a friendly joke
+"research-analyst": use this agent to conduct thorough research on complex topics
+</example_agent_description>
+
+<example>
+user: "Please write a function that checks if a number is prime"
+assistant: Sure let me write a function that checks if a number is prime
+assistant: First let me use the Write tool to write a function that checks if a number is prime
+assistant: I'm going to use the Write tool to write the following code:
+<code>
+function isPrime(n) {
+  if (n <= 1) return false
+  for (let i = 2; i * i <= n; i++) {
+    if (n % i === 0) return false
+  }
+  return true
+}
+</code>
+<commentary>
+Since significant content was created and the task was completed, now use the content-reviewer agent to review the work
+</commentary>
+assistant: Now let me use the content-reviewer agent to review the code
+assistant: Uses the Task tool to launch with the content-reviewer agent
+</example>
+<example>
+user: "Can you help me research the environmental impact of different renewable energy sources and create a comprehensive report?"
+<commentary>
+This is a complex research task that would benefit from using the research-analyst agent to conduct thorough analysis
+</commentary>
+assistant: I'll help you research the environmental impact of renewable energy sources. Let me use the research-analyst agent to conduct comprehensive research on this topic.
+assistant: Uses the Task tool to launch with the research-analyst agent, providing detailed instructions about what research to conduct and what format the report should take
+</example>
+<example>
+user: "Hello"
+<commentary>
+Since the user is greeting, use the greeting-responder agent to respond with a friendly joke
+</commentary>
+assistant: "I'm going to use the Task tool to launch with the greeting-responder agent"
+</example>`;
+
+/**
+ * Description for the edit_file tool
+ * Ported exactly from Python EDIT_DESCRIPTION
+ */
+export const EDIT_DESCRIPTION = `Performs exact string replacements in files.
+Usage:
+
+You must use your Read tool at least once in the conversation before editing. This tool will error if you attempt an edit without reading the file.
+When editing text from Read tool output, ensure you preserve the exact indentation (tabs/spaces) as it appears AFTER the line number prefix. The line number prefix format is: spaces + line number + tab. Everything after that tab is the actual file content to match. Never include any part of the line number prefix in the old_string or new_string.
+ALWAYS prefer editing existing files. NEVER write new files unless explicitly required.
+Only use emojis if the user explicitly requests it. Avoid adding emojis to files unless asked.
+The edit will FAIL if old_string is not unique in the file. Either provide a larger string with more surrounding context to make it unique or use replace_all to change every instance of old_string.
+Use replace_all for replacing and renaming strings across the file. This parameter is useful if you want to rename a variable for instance.`;
+
+/**
+ * Description for the read_file tool
+ * Ported exactly from Python TOOL_DESCRIPTION
+ */
+export const TOOL_DESCRIPTION = `Reads a file from the local filesystem. You can access any file directly by using this tool. Assume this tool is able to read all files on the machine. If the User provides a path to a file assume that path is valid. It is okay to read a file that does not exist; an error will be returned.
+Usage:
+
+The file_path parameter must be an absolute path, not a relative path
+By default, it reads up to 2000 lines starting from the beginning of the file
+You can optionally specify a line offset and limit (especially handy for long files), but it's recommended to read the whole file by not providing these parameters
+Any lines longer than 2000 characters will be truncated
+Results are returned using cat -n format, with line numbers starting at 1
+You have the capability to call multiple tools in a single response. It is always better to speculatively read multiple files as a batch that are potentially useful.
+If you read a file that exists but has empty contents you will receive a system reminder warning in place of file contents.`;

package/src/deep_agent/state.ts

@@ -0,0 +1,63 @@
+/**
+ * State definitions for Deep Agents
+ *
+ * TypeScript equivalents of the Python state classes using LangGraph's Annotation.Root() pattern.
+ * Defines Todo interface and DeepAgentState using MessagesAnnotation as base with proper reducer functions.
+ */
+
+import "@langchain/langgraph/zod";
+import { MessagesZodState } from "@langchain/langgraph";
+import type { Todo } from "./types";
+import { withLangGraph } from "@langchain/langgraph/zod";
+import { z } from "zod";
+
+/**
+ * File reducer function that merges file dictionaries
+ * Matches the Python file_reducer function behavior exactly
+ */
+export function fileReducer(
+  left: Record<string, string> | null | undefined,
+  right: Record<string, string> | null | undefined
+): Record<string, string> {
+  if (left == null) {
+    return right || {};
+  } else if (right == null) {
+    return left;
+  } else {
+    return { ...left, ...right };
+  }
+}
+
+/**
+ * Todo reducer function that replaces the entire todo list
+ * This matches the Python behavior where todos are completely replaced
+ */
+export function todoReducer(
+  left: Todo[] | null | undefined,
+  right: Todo[] | null | undefined
+): Todo[] {
+  if (right != null) {
+    return right;
+  }
+  return left || [];
+}
+
+/**
+ * DeepAgentState using LangGraph's Annotation.Root() pattern
+ * Extends MessagesAnnotation (equivalent to Python's AgentState) with todos and files channels
+ */
+export const DeepAgentState = MessagesZodState.extend({
+  todos: withLangGraph(z.custom<Todo[]>(), {
+    reducer: {
+      schema: z.custom<Todo[]>(),
+      fn: todoReducer,
+    },
+  }),
+
+  files: withLangGraph(z.custom<Record<string, string>>(), {
+    reducer: {
+      schema: z.custom<Record<string, string>>(),
+      fn: fileReducer,
+    },
+  }),
+});

package/src/deep_agent/subAgent.ts

@@ -0,0 +1,185 @@
+/**
+ * SubAgent implementation for Deep Agents
+ *
+ * Task tool creation and sub-agent management.
+ * Creates SubAgent interface matching Python's TypedDict structure and implements
+ * createTaskTool() function that creates agents map, handles tool resolution by name,
+ * and returns a tool function that uses createReactAgent for sub-agents.
+ */
+
+import { tool, StructuredTool } from "@langchain/core/tools";
+import { ToolMessage } from "@langchain/core/messages";
+import {
+  Command,
+  getCurrentTaskInput,
+  GraphInterrupt,
+} from "@langchain/langgraph";
+import { ToolRunnableConfig } from "@langchain/core/tools";
+import { createReactAgent } from "@langchain/langgraph/prebuilt";
+import { z } from "zod";
+import type { LanguageModelLike, SubAgent } from "./types";
+import { writeTodos, readFile, writeFile, editFile, ls } from "./tools";
+import { TASK_DESCRIPTION_PREFIX, TASK_DESCRIPTION_SUFFIX } from "./prompts";
+import { ModelLattice } from "../model_lattice/ModelLattice";
+import { getModelLattice } from "../model_lattice/ModelLatticeManager";
+
+/**
+ * Built-in tools map for tool resolution by name
+ */
+const BUILTIN_TOOLS: Record<string, StructuredTool> = {
+  write_todos: writeTodos,
+  read_file: readFile,
+  write_file: writeFile,
+  edit_file: editFile,
+  ls: ls,
+};
+
+/**
+ * Create task tool function that creates agents map, handles tool resolution by name,
+ * and returns a tool function that uses createReactAgent for sub-agents.
+ * Uses Command for state updates and navigation between agents.
+ */
+export function createTaskTool<
+  StateSchema extends z.ZodObject<any, any, any, any, any>
+>(inputs: {
+  subagents: SubAgent[];
+  tools: Record<string, StructuredTool>;
+  model: ModelLattice;
+  stateSchema: StateSchema;
+}) {
+  const {
+    subagents,
+    tools = {},
+    model = getModelLattice("default")?.client,
+    stateSchema,
+  } = inputs;
+
+  if (!model) {
+    throw new Error("Model not found");
+  }
+
+  // Combine built-in tools with provided tools for tool resolution
+  const allTools = { ...BUILTIN_TOOLS, ...tools };
+
+  // Pre-create all agents like Python does
+  const agentsMap = new Map<string, any>();
+  for (const subagent of subagents) {
+    // Resolve tools by name for this subagent
+    const subagentTools: StructuredTool[] = [];
+    if (subagent.tools) {
+      for (const toolName of subagent.tools) {
+        const resolvedTool = allTools[toolName];
+        if (resolvedTool) {
+          subagentTools.push(resolvedTool);
+        } else {
+          // eslint-disable-next-line no-console
+          console.warn(
+            `Warning: Tool '${toolName}' not found for agent '${subagent.name}'`
+          );
+        }
+      }
+    } else {
+      // If no tools specified, use all tools like Python does
+      subagentTools.push(...Object.values(allTools));
+    }
+
+    // Create react agent for the subagent (pre-create like Python)
+    const reactAgent = createReactAgent({
+      llm: model,
+      tools: subagentTools,
+      stateSchema: stateSchema as any,
+      messageModifier: subagent.prompt,
+      checkpointer: false,
+    });
+
+    agentsMap.set(subagent.name, reactAgent);
+  }
+
+  return tool(
+    async (
+      input: { description: string; subagent_type: string },
+      config: ToolRunnableConfig
+    ) => {
+      const { description, subagent_type } = input;
+
+      // Get the pre-created agent
+      const reactAgent = agentsMap.get(subagent_type);
+      if (!reactAgent) {
+        return `Error: Agent '${subagent_type}' not found. Available agents: ${Array.from(
+          agentsMap.keys()
+        ).join(", ")}`;
+      }
+
+      try {
+        // Get current state for context
+        const currentState = getCurrentTaskInput<z.infer<typeof stateSchema>>();
+
+        // Modify state messages like Python does
+        const modifiedState = {
+          ...currentState,
+          messages: [
+            {
+              role: "user",
+              content: description,
+            },
+          ],
+        };
+
+        // Execute the subagent with the task
+        const result = await reactAgent.invoke(modifiedState, config);
+
+        // Use Command for state updates and navigation between agents
+        // Return the result using Command to properly handle subgraph state
+        return new Command({
+          update: {
+            files: result.files || {},
+            messages: [
+              new ToolMessage({
+                content:
+                  result.messages?.slice(-1)[0]?.content || "Task completed",
+                tool_call_id: config.toolCall?.id as string,
+              }),
+            ],
+          },
+        });
+      } catch (error) {
+        if (error instanceof GraphInterrupt) {
+          throw error;
+        }
+        // Handle errors gracefully
+        const errorMessage =
+          error instanceof Error ? error.message : String(error);
+        return new Command({
+          update: {
+            messages: [
+              new ToolMessage({
+                content: `Error executing task '${description}' with agent '${subagent_type}': ${errorMessage}`,
+                tool_call_id: config.toolCall?.id as string,
+              }),
+            ],
+          },
+        });
+      }
+    },
+    {
+      name: "task",
+      description:
+        TASK_DESCRIPTION_PREFIX.replace(
+          "{other_agents}",
+          subagents.map((a) => `- ${a.name}: ${a.description}`).join("\n")
+        ) + TASK_DESCRIPTION_SUFFIX,
+      schema: z.object({
+        description: z
+          .string()
+          .describe("The task to execute with the selected agent"),
+        subagent_type: z
+          .string()
+          .describe(
+            `Name of the agent to use. Available: ${subagents
+              .map((a) => a.name)
+              .join(", ")}`
+          ),
+      }),
+    }
+  );
+}