# Flowise MCP for AI Agents AI Agent Connect

> Flowise MCP lets you manage your low-code AI workflows directly from your AI client. You can run predictions, check chatflows and agentflows, audit execution logs, and manage credentials without switching tabs. It connects your Flowise instance to your agent so you can build and debug complex AI logic using natural language.

## Overview
- **Category:** ai-frontier
- **Price:** Free
- **Endpoint:** https://edge.vinkius.com/vk_preview_x0CXvywaUNOIkFSe5hI4681xfnamFZZaTdsop8is/ai-agent-connect
- **Tags:** llm-orchestration, rag-pipelines, low-code, ai-agents, workflow-automation, chatflows

## Description

This Flowise MCP connects your FlowiseAI instance to any AI agent, giving you direct control over your low-code generative AI applications through natural conversation. Imagine you're building complex AI agents and find yourself constantly jumping between your code editor and the Flowise dashboard just to check if a node is working or to see why a specific prompt failed. This Connector changes that by bringing your entire Flowise environment into your chat interface. You can ask your agent to pull up specific chatflows, check the status of your agentflows, or run a live prediction to see how the logic handles a new input. Instead of manually digging through logs to find a needle in a haystack, you can just ask for the execution history of a specific flow to see exactly where things went sideways. It also handles the boring stuff like listing your available tools or checking your credentials so you don't have to hunt through settings menus. Because Vinkius makes it easy to connect these types of specialized tools, you can keep your focus on the logic rather than the logistics. It's about having a direct line of sight into your AI orchestration layer while you're actually in the middle of building and testing your work. You can move faster because you're looking at the same data in one place.

## Tools

### list_credentials
This tool lists the credentials stored in your instance. This helps you verify which auth components are currently active.

### get_history
This tool gets the chat execution history for a flow. Use this to pull logs and debug exactly where a logic chain broke.

### list_chatflows
This tool lists all the chatflows currently in your Flowise instance. It helps you quickly see what's deployed.

### get_chatflow
This tool gets the specific details for a single chatflow. Use this to see the architecture of a specific flow.

### predict
This tool runs a live prediction on a chatflow. This lets you test how the AI responds to a specific user question right now.

### list_agentflows
This tool lists your agentflows. This is how you see the complex, multi-step reasoning paths you've built.

### list_tools
This tool lists the tools and integrations in your environment. Use this to see what capabilities your agent can actually call.

## Prompt Examples

**Prompt:** 
```
Show me the logs for the 'SupportBot' flow.
```

**Response:** 
```
Retrieving logs for 'SupportBot'... I found 3 recent executions. The last one failed at the 'Knowledge Retrieval' node with a 'Timeout' error. Would you like to see the full conversation trace?
```

**Prompt:** 
```
Run a test on the 'LeadGen' flow with the prompt: 'How do I sign up?'
```

**Response:** 
```
Prediction started! I've sent that query to the 'LeadGen' flow. Here is the response from the AI: 'To sign up, you can visit our website and click the Register button. Would you like me to send you the link?'
```

**Prompt:** 
```
What tools does my Flowise instance have available?
```

**Response:** 
```
I've checked your Flowise instance. You currently have 8 tools connected, including: 
* Google Calendar
* Slack Messaging
* Stripe Payments
* Internal CRM
* Web Search
* PDF Parser
* Email Sender
* Database Query
```

## Capabilities

### Run live AI predictions
Submit questions to chatflows and get responses in real-time.

### Check execution logs
Pull past traces and conversational logs to debug logic.

### Manage chatflows
List and retrieve details for all deployed flows.

### Monitor agentflows
Access compound flows and multi-step reasoning logic.

### Check credentials
List stored auth components for your logic chains.

### Discover tools
Retrieve custom tools and third-party integrations.

## Use Cases

### Debugging a broken chain
An engineer asks the agent to show the history for a specific flow to see why a RAG step failed.

### Testing a new prompt
A developer uses predict to see how a new system instruction changes the output of a customer support bot.

### Inventory check
A product manager asks to list all chatflows to see which ones are live in the production environment.

### Tool verification
An automation lead asks to list tools to see if the agent has access to the latest Slack integration.

## Benefits

- Debug faster by pulling execution history logs directly into your chat without leaving your workspace.
- Test your logic instantly by using predict to see real-time AI responses to specific user prompts.
- Keep your environment organized by using list_chatflows to see every deployed flow in one list.
- Verify your integrations by using list_tools to see every custom action your agent can perform.
- Simplify your dev cycle by using list_agentflows to manage complex multi-step reasoning paths.
- Secure your setup by using list_credentials to check which auth components are currently active.

## How It Works

The bottom line is you get a direct command line to your Flowise instance through your favorite AI chat interface.

1. Subscribe to the Connector and grab your Flowise Base URL and API Key from your settings.
2. Connect the Connector to your AI client by providing those credentials.
3. Start asking your agent to run predictions, list flows, or audit logs.

## Frequently Asked Questions

**What is the Flowise MCP?**
It's a way to connect your Flowise AI instance to your AI client so you can manage workflows and run tests using natural language.

**Can I use Flowise MCP to run live tests?**
Yes, you can use the predict tool to send a prompt to any of your chatflows and see the AI's response instantly.

**How do I see my execution logs with Flowise MCP?**
You can ask your agent to pull the history for any flow, and it will show you the past traces and conversational logs.

**Does Flowise MCP support agentflows?**
Yes, it allows you to access and manage compound agentflows that handle multi-step reasoning and complex tasks.

**Can I check my credentials using Flowise MCP?**
You can use it to list the credentials stored in your Flowise instance to verify which auth components are active.

**Is Flowise MCP good for debugging?**
It's great for debugging because it lets you pull specific logs and see why a logic chain failed without leaving your chat.

**How do I connect Flowise MCP to my agent?**
You just need to provide your Flowise Base URL and API Key in your AI client settings to get started.

**Can my agent run a prediction against a specific Flowise chatflow?**
Yes. Use the 'predict' tool. Provide the 'chatflow_id' and your question. The agent will command the Flowise backend to process the logic chain and return the AI-generated response directly in your chat.

**How do I see the past conversational logs for a chatflow via chat?**
Use the 'get_history' tool with the 'chatflow_id'. Your agent will retrieve the past execution traces and logs, helping you understand how users have interacted with that specific logic chain natively.

**Can I list all custom tools configured in my Flowise instance through the agent?**
Absolutely. Use the 'list_tools' tool. Your agent will retrieve custom tools and integrations configured in your environment, allowing you to audit available capabilities through natural conversation.