diff --git a/src/tools/descriptions/Task.md b/src/tools/descriptions/Task.md
index 02ca1e7..a9727ff 100644
--- a/src/tools/descriptions/Task.md
+++ b/src/tools/descriptions/Task.md
@@ -6,12 +6,22 @@ The Task tool launches specialized agents (subprocesses) that autonomously handl
 
 ## Usage
 
-When using the Task tool, you must specify:
+The Task tool supports two commands:
+
+### Run (default)
+Launch a subagent to perform a task. Parameters:
 - **subagent_type**: Which specialized agent to use (see Available Agents section)
 - **prompt**: Detailed, self-contained instructions for the agent (agents cannot ask questions mid-execution)
 - **description**: Short 3-5 word summary for tracking
 - **model** (optional): Override the model for this agent
 
+### Refresh
+Re-scan the `.letta/agents/` directories to discover new or updated custom subagents:
+```typescript
+Task({ command: "refresh" })
+```
+Use this after creating or modifying custom subagent definitions.
+
 ## When to use this tool:
 
 - **Codebase exploration**: Use when you need to search for files, understand code structure, or find specific patterns
diff --git a/src/tools/impl/Task.ts b/src/tools/impl/Task.ts
index 967c077..66b5ecc 100644
--- a/src/tools/impl/Task.ts
+++ b/src/tools/impl/Task.ts
@@ -5,7 +5,11 @@
  * Supports both built-in subagent types and custom subagents defined in .letta/agents/.
  */
 
-import { getAllSubagentConfigs } from "../../agent/subagents";
+import {
+  clearSubagentConfigCache,
+  discoverSubagents,
+  getAllSubagentConfigs,
+} from "../../agent/subagents";
 import { spawnSubagent } from "../../agent/subagents/manager";
 import {
   completeSubagent,
@@ -15,9 +19,10 @@ import {
 import { validateRequiredParams } from "./validation";
 
 interface TaskArgs {
-  subagent_type: string;
-  prompt: string;
-  description: string;
+  command?: "run" | "refresh";
+  subagent_type?: string;
+  prompt?: string;
+  description?: string;
   model?: string;
   toolCallId?: string; // Injected by executeTool for linking subagent to parent tool call
   signal?: AbortSignal; // Injected by executeTool for interruption handling
@@ -27,15 +32,45 @@ interface TaskArgs {
  * Task tool - Launch a specialized subagent to handle complex tasks
  */
 export async function task(args: TaskArgs): Promise<string> {
-  // Validate required parameters
+  const { command = "run", model, toolCallId, signal } = args;
+
+  // Handle refresh command - re-discover subagents from .letta/agents/ directories
+  if (command === "refresh") {
+    // Clear the cache to force re-discovery
+    clearSubagentConfigCache();
+
+    // Discover subagents from global and project directories
+    const { subagents, errors } = await discoverSubagents();
+
+    // Get all configs (builtins + discovered) to report accurate count
+    const allConfigs = await getAllSubagentConfigs();
+    const totalCount = Object.keys(allConfigs).length;
+    const customCount = subagents.length;
+
+    // Log any errors
+    if (errors.length > 0) {
+      for (const error of errors) {
+        console.warn(
+          `Subagent discovery error: ${error.path}: ${error.message}`,
+        );
+      }
+    }
+
+    const errorSuffix = errors.length > 0 ? `, ${errors.length} error(s)` : "";
+    return `Refreshed subagents list: found ${totalCount} total (${customCount} custom)${errorSuffix}`;
+  }
+
+  // For run command, validate required parameters
   validateRequiredParams(
     args,
     ["subagent_type", "prompt", "description"],
     "Task",
   );
 
-  const { subagent_type, prompt, description, model, toolCallId, signal } =
-    args;
+  // Extract validated params (guaranteed to exist after validateRequiredParams)
+  const subagent_type = args.subagent_type as string;
+  const prompt = args.prompt as string;
+  const description = args.description as string;
 
   // Get all available subagent configs (built-in + custom)
   const allConfigs = await getAllSubagentConfigs();
diff --git a/src/tools/schemas/Task.json b/src/tools/schemas/Task.json
index 4ef705b..738ab62 100644
--- a/src/tools/schemas/Task.json
+++ b/src/tools/schemas/Task.json
@@ -1,24 +1,28 @@
 {
   "type": "object",
   "properties": {
+    "command": {
+      "type": "string",
+      "enum": ["run", "refresh"],
+      "description": "The operation to perform: \"run\" to spawn a subagent (default), \"refresh\" to re-scan the .letta/agents/ directories and update the available subagents list"
+    },
     "subagent_type": {
       "type": "string",
-      "description": "The type of specialized agent to use. Available agents are discovered from .letta/agents/ directory."
+      "description": "The type of specialized agent to use. Available agents are discovered from .letta/agents/ directory. Required for \"run\" command."
     },
     "prompt": {
       "type": "string",
-      "description": "The task for the agent to perform"
+      "description": "The task for the agent to perform. Required for \"run\" command."
     },
     "description": {
       "type": "string",
-      "description": "A short (3-5 word) description of the task"
+      "description": "A short (3-5 word) description of the task. Required for \"run\" command."
     },
     "model": {
       "type": "string",
       "description": "Optional model to use for this agent. If not specified, uses the recommended model for the subagent type."
     }
   },
-  "required": ["subagent_type", "prompt", "description"],
   "additionalProperties": false,
   "$schema": "http://json-schema.org/draft-07/schema#"
 }