# Poe MCP for AI Agents AI Agent Connect

> Poe MCP lets you manage your chatbot fleet directly from your AI client. You can create, update, and delete bots, query other models for complex workflows, and pull real-time usage stats like message counts and response times. It turns your AI into a powerhouse for bot development and multi-model orchestration.

## Overview
- **Category:** ai-frontier
- **Price:** Free
- **Endpoint:** https://edge.vinkius.com/vk_preview_GTjMlzgwII8zcah2W5vaPUuxivd67B9v9CRpL0LG/ai-agent-connect
- **Tags:** chatbot-development, llm-orchestration, model-chaining, ai-platforms, bot-management, conversational-ai

## Description

Managing a fleet of chatbots usually means jumping between different dashboards, manually copying API keys, and squinting at logs to see if your bots are actually working. This Poe MCP changes that by bringing the entire bot lifecycle into your chat interface. You can build out complex multi-model workflows where your bot calls other models to handle specific tasks, all while keeping an eye on the numbers that matter. Instead of hunting for data, you just ask your agent to show you the latest message counts or error rates for a specific bot. It's a huge win for anyone building on Quora's platform because it treats your bot fleet like a programmable database. Since Vinkius makes it easy to connect these types of tools, you can move from idea to live bot in a few prompts. You get to focus on the logic of your bots rather than the friction of the platform.

## Tools

### create_bot
Create a new API bot on Poe with a name, model, and system prompt. This helps you spin up new bots instantly.

### delete_bot
Permanently remove a Poe bot and its entire conversation history. Use this to clean up old projects.

### get_bot_stats
Fetch message counts, unique users, and response times for a bot. Use this to monitor your bot's health.

### get_bot
Retrieve the specific details and configuration of a single Poe bot. This is great for checking current settings.

### list_bots
Get a full list of bot names, handles, and statuses under your account. Use this to see your whole fleet.

### list_messages
Pull recent conversation logs to debug responses or analyze user patterns. This helps you see what users are asking.

### query_bot
Call another Poe bot from your own bot to chain model responses together. This lets you build complex AI logic.

### send_message
Simulate a user interaction to test your bot's connectivity and response. This verifies your bot is live.

### update_bot
Change a bot's configuration and system prompt immediately. This lets you iterate on bot behavior quickly.

### list_available_bots
See which bots are available for your bot to query in a chain. This is essential for multi-model workflows.

## Prompt Examples

**Prompt:** 
```
List my bots and show me the message count for the one called ResearchBot.
```

**Response:** 
```
🤖 You have 4 bots in your fleet. Here are the stats for **ResearchBot**:

| Metric | Value |
| :--- | :--- |
| **Total Messages** | 1,247 |
| **Unique Users** | 89 |
| **Avg. Response Time** | 2.3s |

Would you like to see the messages for any other bots?
```

**Prompt:** 
```
Create a new bot named 'Travel Guide' using the Claude model.
```

**Response:** 
```
✅ **Bot Created Successfully!**

*   **Name:** Travel Guide
*   **Model:** Claude
*   **Status:** Active

Your new bot is ready. You can now query it or update its system prompt whenever you like.
```

**Prompt:** 
```
Query the 'Translator' bot from my 'Assistant' bot with the phrase 'Hello'.
```

**Response:** 
```
🤖 **Chaining Response Received:**

Your 'Assistant' bot successfully queried the 'Translator' bot. 

**Input:** 'Hello'
**Translator Output:** 'Bonjour'

The response was passed back to your main assistant successfully.
```

## Capabilities

### Create new bots
Build new API bots with specific names, models, and system prompts.

### Chain model responses
Have your bot query other models on Poe to handle complex tasks.

### Pull usage stats
Get real-time data on message counts, unique users, and response times.

### Monitor conversations
View recent user messages to debug responses and analyze patterns.

### Update bot settings
Change configuration and prompts for your existing bots on the fly.

### Test connectivity
Send test messages to verify that your bot is responding correctly.

## Use Cases

### Debugging a non-responsive bot
A developer notices a bot is acting up. They ask their agent to use send_message to see the exact response from the endpoint.

### Comparing model outputs
A researcher uses query_bot to have their main bot call two different models and compare the results in a single chat.

### Bulk updating bot prompts
A creator needs to change the behavior of a bot. They tell their agent to use update_bot to modify the system prompt instantly.

### Analyzing user trends
A community manager asks the agent to list_messages for their top bot to see what common questions users are asking today.

## Benefits

- Stop switching tabs because you can manage your entire Poe bot fleet from one chat window.
- Build complex logic by using query_bot to chain different models into a single workflow.
- Get instant feedback on bot health using get_bot_stats to see message counts and errors.
- Debug conversations faster by pulling recent logs with list_messages directly into your workspace.
- Rapidly iterate on bot behavior by updating system prompts on the fly with update_bot.
- Test your bot's live connectivity instantly using send_message without manual clicking.

## How It Works

The bottom line is you get a unified control panel for your Poe bot fleet inside your favorite AI client.

1. Get your API access token from the Poe creator dashboard.
2. Add the Poe MCP to your client and provide the token.
3. Start managing your bots or querying models through natural language.

## Frequently Asked Questions

**Can I use Poe MCP to manage my bots in Cursor?**
Yes. You can connect this Connector to Cursor to manage your Poe bots, check stats, and update prompts without leaving your code editor.

**Does Poe MCP let me see how many people used my bot?**
Yes, it pulls real-time usage statistics. You can ask your agent to show you message counts and unique user numbers for any bot in your account.

**How do I use Poe MCP to chain different models?**
You can use the query tool to have one bot call another. This allows you to build complex workflows where different models handle different parts of a task.

**Can I update my bot's prompt without going to the website?**
Yes. You can use the update tool to change a bot's configuration or system prompt directly through your AI client's chat interface.

**How do I test if my Poe bot is actually responding?**
You can ask your agent to send a test message to your bot. It will simulate a user interaction and show you the response your bot generates.

**Does Poe MCP work for both personal and creator accounts?**
It works for any account that has a Poe API access token from the creator dashboard. This includes both personal and professional accounts.

**How do I get a Poe API access token?**
Go to [creator.poe.com](https://creator.poe.com/), navigate to Settings > API, and generate an access token. This token authenticates all API requests for bot management.

**What AI models can I query through Poe?**
Poe has GPT-4, GPT-4o, Claude 3.5 Sonnet, Claude 3 Opus, Llama 3, and dozens of specialized bots. Use query_bot to chain responses from multiple models in your workflows.

**How does bot chaining work?**
Using Poe API v2, your bot can query any other bot on Poe for free. The response becomes input for your bot's processing. Costs are covered by the user's free message limit or subscription.