List Prompts

Discovers all available prompts on an MCP server. This node returns a list of prompts with their names, descriptions, and parameters, enabling dynamic prompt discovery.

Common Properties

• Name - The custom name of the node.
• Color - The custom color of the node.
• Delay Before (sec) - Waits in seconds before executing the node.
• Delay After (sec) - Waits in seconds after executing node.
• Continue On Error - Automation will continue regardless of any error. The default value is false.

Inputs

• Client Id - The MCP client connection identifier from the Connect node.

Options

• Timeout - Timeout in seconds for the operation (default: 30).

Outputs

• Prompts - Array of prompt objects, each containing name, description, and arguments (if any).

How It Works

1. Client Lookup - Retrieves the MCP client using the provided Client ID
2. List Request - Sends a ListPromptsRequest to the MCP server
3. Response Processing - Extracts prompt information from the server response
4. Array Creation - Converts prompts to an array of objects with name, description, and arguments
5. Output - Returns the prompts array

Requirements

• Valid Client ID from a Connect node
• Active connection to the MCP server

Output Format

Each prompt in the array contains:

{
  "name": "prompt_name",
  "description": "What the prompt does",
  "arguments": [
    {
      "name": "param_name",
      "description": "Parameter description",
      "required": true
    }
  ]
}

The arguments field is optional and only present if the prompt accepts parameters.

Examples

Example 1: List All Prompts

Inputs:

• Client Id: prompt-server

Prompts Output:

[
  {
    "name": "greeting",
    "description": "Generate a personalized greeting message"
  },
  {
    "name": "code_review",
    "description": "Generate a code review prompt",
    "arguments": [
      {
        "name": "language",
        "description": "Programming language",
        "required": true
      },
      {
        "name": "focus",
        "description": "Review focus area",
        "required": false
      }
    ]
  },
  {
    "name": "email_template",
    "description": "Generate email template",
    "arguments": [
      {
        "name": "recipient",
        "description": "Recipient name",
        "required": true
      },
      {
        "name": "subject",
        "description": "Email subject",
        "required": true
      }
    ]
  }
]

---

Example 2: Process Prompt List

Flow:

Connect → List Prompts → Loop through prompts → Display or store

JavaScript processing:

// Access prompts from List Prompts output
const prompts = msg.prompts;

// Extract just prompt names
msg.prompt_names = prompts.map(p => p.name);

// Filter prompts that require parameters
msg.parameterized_prompts = prompts.filter(p => p.arguments && p.arguments.length > 0);

// Find specific prompt
msg.code_review_prompt = prompts.find(p => p.name === 'code_review');

---

Example 3: Dynamic Prompt Selection

Flow:

List Prompts → Filter by criteria → Select best match → Get Prompt

Example:

// Find prompts related to email
const emailPrompts = msg.prompts.filter(p =>
  p.name.includes('email') || p.description.toLowerCase().includes('email')
);

// Use the first matching prompt
if (emailPrompts.length > 0) {
  msg.selected_prompt = emailPrompts[0].name;
}

---

Example 4: Build Prompt Catalog

Flow:

Connect → List Prompts → Format catalog → Save to database

Create documentation:

const catalog = msg.prompts.map(prompt => ({
  name: prompt.name,
  description: prompt.description,
  required_params: prompt.arguments?.filter(a => a.required).map(a => a.name) || [],
  optional_params: prompt.arguments?.filter(a => !a.required).map(a => a.name) || []
}));

msg.prompt_catalog = catalog;

Best Practices

1. Discovery:

   • List prompts at the start of automation to know what's available
   • Cache the list if prompts don't change frequently
   • Use to validate prompt names before calling Get Prompt

2. Processing:

   • Filter prompts by name patterns or descriptions
   • Check for required vs optional parameters
   • Build dynamic UIs or catalogs from the list

3. Error Handling:

   • Check if prompts array is not empty
   • Handle servers with no prompts gracefully
   • Validate server connection before listing

4. Performance:

   • Cache prompt lists to reduce server calls
   • Only refresh when needed
   • Use for initialization, not every iteration

Common Errors and Solutions

Error: "Client Id cannot be empty"

Solution:

• Ensure Connect node executed successfully
• Pass Client Id from Connect node
• Verify client connection is active

Error: "Failed to list prompts"

Solution:

• Check server is running and accessible
• Verify Client Id is correct
• Check server logs for errors
• Ensure server has ListPrompts capability enabled

Empty Prompts Array

Cause: Server has no prompts registered.

Solution:

• Verify server has prompts configured
• Check server is properly initialized
• Ensure prompts are registered correctly on server

Tips for Effective Use

1. Initial Discovery - Use at flow start to discover capabilities
2. Validation - Validate prompt names exist before Get Prompt
3. Documentation - Generate prompt documentation automatically
4. Dynamic Selection - Select prompts based on context
5. Caching - Store prompt list to avoid repeated calls
6. Filtering - Filter by name patterns or required parameters

Integration Patterns

Pattern 1: Prompt Discovery and Usage

Connect → List Prompts → Select Prompt → Get Prompt → Use

Pattern 2: Validation Before Use

List Prompts → Check if exists → Get Prompt
                                ↓ (if not exists)
                              Error Handler

Pattern 3: Build Catalog

List Prompts → Format → Store in Database → Display to Users

Pattern 4: Dynamic Menu

List Prompts → Create Menu → User Selects → Get Prompt

Processing Examples

Extract Prompt Names Only

const promptNames = msg.prompts.map(p => p.name);
// ["greeting", "code_review", "email_template"]

Find Prompts with Parameters

const parametric = msg.prompts.filter(p =>
  p.arguments && p.arguments.length > 0
);

Create Lookup Map

const promptMap = {};
msg.prompts.forEach(p => {
  promptMap[p.name] = {
    description: p.description,
    params: p.arguments || []
  };
});
msg.prompt_lookup = promptMap;

Check if Specific Prompt Exists

const hasCodeReview = msg.prompts.some(p => p.name === 'code_review');
if (!hasCodeReview) {
  throw new Error('Required prompt not available');
}