A powerful NLU tool for classifying text into custom intents with confidence scoring and entity extraction.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install intent-router
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install intent-router using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Intent Router?
Intent Router is a high-performance Natural Language Understanding (NLU) utility designed to streamline how developers manage user input and agent orchestration. As a key component in the Openclaw Skills ecosystem, it leverages the x402 engine to provide robust text classification without the need for traditional API keys, making it a privacy-conscious and efficient choice for modern AI workflows.
This skill is particularly effective for developers building complex multi-agent systems where accurate message routing is essential. By integrating this tool into your Openclaw Skills library, you can automate the process of determining user intent, extracting critical entities like locations or dates, and ensuring that every request reaches the correct handler with high confidence.
Intent Router Use Cases
- Routing user messages to specific AI agents or microservices based on identified intent.
- Enhancing chatbot interactions by identifying specific entities like dates, locations, and values.
- Implementing automated customer support triage by categorizing tickets as feedback, complaints, or questions.
- Powering multi-language NLU applications within the Openclaw Skills framework.
How Intent Router Works
- The user provides a raw text string and a list of candidate intents to the intent-router.
- The skill processes the text using the x402 engine, analyzing semantics, syntax, and context.
- A primary intent is identified along with a confidence score and a detailed reasoning string to explain the logic.
- Relevant entities (e.g., names, locations, timeframes) are extracted from the text content.
- The system returns a structured JSON response containing the classification data for further routing or processing in your Openclaw Skills workflow.
Intent Router Setup
To get started with this skill from the Openclaw Skills collection, ensure you have the necessary CLI tools installed:
npm install -g [email protected]
awal auth login
Once authenticated, you can invoke the router using the following command structure:
npx [email protected] x402 pay \
https://anicca-proxy-production.up.railway.app/api/x402/intent-router \
-X POST \
-d '{"text":"I want to book a flight to Tokyo next week","intents":["booking","complaint","question"],"language":"en"}'
Intent Router Data Schema & Taxonomy
The intent-router utilizes a structured schema to ensure compatibility across different Openclaw Skills integrations.
Input Parameters
| Field | Type | Description |
|---|---|---|
| text | String | The input text to be analyzed (max 2000 chars). |
| intents | Array | A list of 2-20 candidate intents for classification. |
| language | Enum | Optional language hint (en, ja, es, fr, de, zh, ko). |
| context | String | Optional additional context to improve accuracy. |
Output Response
| Key | Description |
|---|---|
| matched_intent | The most likely intent detected from the candidate list. |
| confidence | A numerical score (0-1) representing classification certainty. |
| entities | An array of detected objects containing type and value. |
| reasoning | A natural language explanation of the classification logic. |
Intent Router Advanced Features
- Zero API key requirement utilizing the x402 payment protocol for enhanced privacy and cost-efficiency.
- Multi-language support with automatic language detection capabilities across seven major languages.
- Secondary intent classification to handle ambiguous or multi-faceted user requests within Openclaw Skills.
- Native entity extraction for capturing variables like dates, locations, and custom values without additional processing.
- Low-latency message routing optimized for real-time AI agent orchestration.
SKILL.md
Loading
Related Openclaw Skills
focus-coach
A behavior-design diagnostic tool that uses the BJ Fogg B=MAP framework to identify focus blockers and prescribe immediate, tiny actions.
Emotion Detector
A sophisticated emotional analysis tool for AI agents that identifies user sentiment and provides actionable response strategies.
Daily Questions
A daily interactive routine that refines AI agent behavior and user profiles through scheduled Telegram questionnaires.
Addis Assistant
A specialized linguistic skill for high-accuracy Amharic speech-to-text transcription and text translation via the Addis Assistant API.
Prompt Sanitizer
An AI-powered security layer that sanitizes prompts by removing PII and detecting malicious injection attempts before they reach your LLM.
Tone Rewriter
A sophisticated linguistic utility that programmatically adapts text into ten distinct tones while preserving original meaning.
Featured*
