> ## Documentation Index
> Fetch the complete documentation index at: https://praison.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Router Agent CLI

> CLI commands for router agent in PraisonAI TypeScript

The `praisonai-ts` CLI provides the `router` command for request routing analysis.

```mermaid theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
graph LR
    User([User]) --> CLI[CLI]
    CLI --> Router([Route])

    classDef agent fill:#8B0000,stroke:#7C90A0,color:#fff
    classDef tool fill:#189AB4,stroke:#7C90A0,color:#fff

    class Router agent
    class User,CLI tool
    classDef agent fill:#8B0000,color:#fff
    classDef tool fill:#189AB4,color:#fff
```

## Quick Start

**1. Simple Usage**

```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
# Analyze input and suggest routing
praisonai-ts router analyze "Help me debug this code"

# Get JSON output
praisonai-ts router analyze "What is 2+2?" --json
```

**Example Output:**

```json theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
{
  "success": true,
  "data": {
    "input": "Help me debug this code",
    "selectedRoute": "code",
    "confidence": 0.95,
    "matchedRoutes": ["code", "question", "general"]
  }
}
```

## Route Categories

The router analyzes input and suggests appropriate routing:

| Route | Triggers |
| - | - |
| `code` | code, debug, function, programming |
| `math` | calculate, math, equation |
| `research` | research, find, search |
| `creative` | write, create, generate |
| `question` | what, how, why, explain |
| `greeting` | hello, hi, hey |
| `general` | Default fallback |

## SDK Usage

For programmatic routing:

```typescript theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
import { RouterAgent, Agent } from 'praisonai';

const codeAgent = new Agent({ name: 'CodeExpert', instructions: '...' });
const mathAgent = new Agent({ name: 'MathExpert', instructions: '...' });

const router = new RouterAgent({
  routes: [
    { agent: codeAgent, condition: (input) => /code|debug/i.test(input) },
    { agent: mathAgent, condition: (input) => /math|calculate/i.test(input) }
  ]
});

const result = await router.route('Help me debug this');
```

For more details, see the [Router Agent SDK documentation](https://praison.ai/docs/js/router-agent.md).

***

## Related

> [card] [Router Agent SDK](https://praison.ai/docs/js/router-agent.md)
>
> Programmatic routing

> [card] [Routing](https://praison.ai/docs/js/routing.md)
>
> Request routing


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.