Skip to main content

​
Overview

FlowiseAI is a visual workflow builder for AI agents with native MCP support.

​
Prerequisites

​
Setup Steps

​
Step 1: Prepare StackOne Configuration

First, prepare your StackOne connection details:
  1. Get your API key from the StackOne Dashboard
  2. Get your account ID from the Accounts section
  3. Encode your API key for Basic auth:
echo -n "<stackone_api_key>" | base64
Save this base64 token - you’ll need it for configuration.

​
Step 2: Add MCP Server to Flowise

1

Open Flowise

Navigate to your Flowise instance (typically http://localhost:3000)
2

Create or Open a Chatflow

Create a new chatflow or open an existing one where you want to add StackOne tools
3

Add MCP Tool

  1. In the left sidebar, find “MCP Tools” or “Tools” section
  2. Drag the MCP Tool component onto your canvas
  3. Connect it to your agent or chat model
4

Configure MCP Server

In the MCP Tool configuration panel:Server Type: Select “HTTP” (StackOne uses HTTP transport)Server URL:
https://api.stackone.com/mcp
Headers: Click “Add Header” and configure:
{
  "Authorization": "Basic <YOUR_BASE64_TOKEN>",
  "x-account-id": "<YOUR_ACCOUNT_ID>",
  "Content-Type": "application/json",
  "Accept": "application/json,text/event-stream"
}
Protocol Version: 2025-06-18 (optional)
5

Test Connection

Click “Connect” or “Test” to verify the MCP server connectionYou should see available tools populate from your StackOne account

​
Step 3: Configure Available Tools

Once connected, Flowise will discover all available StackOne tools for your account: Tool Filtering:
  • You can filter which tools to expose to your agent
  • Search for specific operations by provider or action type
  • Enable/disable tools as needed for your workflow
Example Tools Based on Your Integrations:
  • gong_crm_search_calls - Search CRM calls in Gong
  • salesforce_list_accounts - List accounts from Salesforce
  • slack_send_message - Send messages via Slack
  • Plus hundreds more based on your connected integrations
To see all available integrations and operations:

​
Step 4: Build Your Workflow

Now you can use StackOne tools in your Flowise chatflow:
  1. Connect to Agent: Link the MCP Tool to your AI agent node
  2. Configure Agent Prompt: Tell your agent when to use StackOne tools
  3. Add Logic: Combine with other Flowise components (conditional logic, data transformation, etc.)
  4. Test: Try asking questions that require StackOne data
Example Agent Prompt: 
You are a business data assistant with access to multiple platforms.
Use the appropriate tools based on what the user asks for.
You have access to data from connected business systems and platforms.
Always provide clear, helpful responses based on the data from connected systems.

​
Example Workflows

​
Workflow 1: Cross-Platform Search Assistant

Components:
  1. Chat Input
  2. MCP Tool (StackOne - multiple providers)
  3. Chat Model (GPT-4, Claude, etc.)
  4. Chat Output
Use Case: Users can search across multiple connected platforms from a single interface

​
Workflow 2: Automated Activity Dashboard

Components:
  1. Schedule Trigger (daily)
  2. MCP Tool (StackOne - multiple providers)
  3. Data Aggregation
  4. Notification (Slack, Email)
Use Case: Automatically fetch activity from multiple systems and send daily summaries

​
Workflow 3: Multi-System Data Agent

Components:
  1. Chat Input
  2. MCP Tool (StackOne with multiple accounts/providers)
  3. MCP Tool (Other services)
  4. Chat Model with function calling
  5. Chat Output
Use Case: Agent that can pull and correlate data across your entire business platform ecosystem

​
Troubleshooting

​
Connection Issues

Check:
  • URL is exactly https://api.stackone.com/mcp
  • Base64 token includes the colon: api_key:
  • Account ID is correct
  • Headers are properly formatted JSON
Test with cURL:
curl -X POST https://api.stackone.com/mcp \
  -H 'Authorization: Basic <YOUR_TOKEN>' \
  -H 'x-account-id: <YOUR_ACCOUNT_ID>' \
  -H 'Content-Type: application/json' \
  -d '{"jsonrpc":"2.0","id":"1","method":"tools/list"}'
Possible causes:
  • No active integrations in your StackOne account
  • Incorrect account ID
  • API key doesn’t have proper permissions
Verify: Log into StackOne Dashboard and check:
  • Accounts section shows linked integrations
  • API key is valid and not expired
Common issues:
  • Missing required parameters
  • Provider connection issue
  • Rate limiting
Check:
  • Flowise logs for error messages
  • StackOne Request Logs for API errors
  • Tool schema matches expected parameters
StackOne uses HTTP transport only, not SSE (Server-Sent Events).Configuration:
  • Select “HTTP” as transport type
  • Do NOT use SSE endpoints
  • URL should be https://api.stackone.com/mcp without /sse suffix

​
Best Practices

​
Performance

  1. Tool Selection: Only enable tools you need in your workflow
  2. Parallel Execution: Flowise can execute multiple tools in parallel

​
Security

  1. Environment Variables: Store credentials in environment variables, not hardcoded
  2. Access Control: Use Flowise’s access control features
  3. API Key Rotation: Rotate StackOne API keys periodically

​
Limitations

  • Transport: StackOne only supports HTTP transport (no STDIO or SSE)
  • Session Management: StackOne MCP is stateless (no session persistence)
  • Tool Discovery: Tools are determined by the enabled actions for the integration configuration associated with the linked account
  • Rate Limits: Subject to StackOne API rate limits (60 requests/min by default)

​
Next Steps

Explore All Tools

See all available operations in the StackOne Playground

API Documentation

Learn about StackOne’s Platform API

FlowiseAI Docs

Official Flowise MCP documentation

Get Support

Contact StackOne support for assistance

​
Additional Resources