# FlowCastle

FlowCastle lets an AI agent build and edit Telegram bot flows in a visual workspace. Connect the MCP server to turn a written request into flow changes.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 2
- Author: FlowCastle
- License: MIT · policy=link-only · description only; the file stays with the author
- Source: https://github.com/FlowCastle/telegram-bot-templates
- Updated: 2026-07-25
- HTML: https://affpapa.org/en/nevibecode/mcp/flowcastle

## What it does

The agent can inspect a FlowCastle workspace, prepare editor actions, validate them, and apply them. For instance, a request can add a `/start` greeting, button questions, contact variables, or a notification. You can also view and edit the resulting blocks in FlowCastle’s visual builder.

The repository includes bot templates in JSON. An agent can search for an example, retrieve it, and apply it to a workspace; the files also show how flow steps are arranged. The MCP server uses HTTP. You need a FlowCastle application key beginning with `app_` for authorization, so the workspace cannot be connected without that credential.

In Claude Code, add the server with `claude mcp add --transport http flowcastle https://api.flowcastle.ai/api/mcp`. Codex uses a separate command. FlowCastle is also listed for Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, and Gemini CLI.

## Requirements
- Remote server: `https://api.flowcastle.ai/api/mcp`

## Install

### Claude Code (native platform)

**Platform CLI: remote server**
```bash
claude mcp add --transport http flowcastle https://api.flowcastle.ai/api/mcp
```

### Codex

**Platform CLI: remote server**
```bash
codex mcp add flowcastle --url https://api.flowcastle.ai/api/mcp
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "flowcastle": {
            "url": "https://api.flowcastle.ai/api/mcp"
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "flowcastle": {
            "type": "http",
            "url": "https://api.flowcastle.ai/api/mcp"
        }
    }
}
```

### Gemini CLI

**Platform CLI: remote server**
```bash
gemini mcp add -t http flowcastle https://api.flowcastle.ai/api/mcp
```

## When to use it
- create a Telegram bot flow from a written brief
- store answers to button questions in contact variables
- validate flow actions before applying them
- find a bot template and adapt it to a workspace

## Example prompts
- Add a `/start` greeting and button questions to my FlowCastle workspace; save the replies in contact variables.
- Find a Telegram support bot template and apply it to my workspace.
- Validate the actions for this flow, then apply them if they pass.

## Similar
- [claude-ops-telegram-server](https://affpapa.org/en/nevibecode/mcp/claude-ops-telegram-server.md) — MCP server · ★ 547
- [Notify MCP Server](https://affpapa.org/en/nevibecode/mcp/notify-mcp-server.md) — MCP server · ★ 29
- [TONNode — TON blockchain access](https://affpapa.org/en/nevibecode/mcp/tonnode-ton-blockchain-access.md) — MCP server · ★ 27
- [telegram-mcp](https://affpapa.org/en/nevibecode/mcp/telegram-mcp.md) — MCP server · ★ 7

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md