| name | Text Classifier |
| slug | text-classifier |
| description | Classify text into predefined or custom categories. Use when agents need to route tickets, moderate content, categorize emails, or tag documents. Uses keyword matching and TF-IDF scoring. No external API needed.
|
| category | NLP |
| tags | ["classification","nlp","text-analysis","categorization","content-moderation"] |
| price_per_call | 0 |
| input_schema | {"type":"object","properties":{"text":{"type":"string","description":"Text to classify (10-50,000 chars)"},"categories":{"type":"object","description":"Optional custom categories map {category: [keywords]}. Uses predefined if omitted."}},"required":["text"]} |
| output_schema | {"type":"object","properties":{"category":{"type":"string","description":"Top predicted category"},"confidence":{"type":"number","description":"Confidence score 0-100"},"top_categories":{"type":"array","description":"Top 3 categories with scores"}}} |
| metadata | {"requires":{"env":["CLAW0X_API_KEY"]}} |
Text Classifier
Classify text into categories using keyword matching and TF-IDF scoring. Supports 8 predefined categories (technology, business, science, health, sports, politics, entertainment, education) or custom categories you define.
How It Works
- Tokenize input text, remove stop words
- Match against category keyword lists
- Score using term frequency relative to text length
- Return top category with confidence and ranked alternatives
Use Cases
- Content moderation (flag inappropriate content)
- Ticket routing (assign support tickets to teams)
- Email categorization (inbox triage)
- Document tagging (auto-label content)
- News classification (topic detection)
Prerequisites
- Sign up at claw0x.com
- Create API key in Dashboard
- Set environment variable:
export CLAW0X_API_KEY="ck_live_..."
Pricing
FREE. No charge per call.
- Requires Claw0x API key for authentication
- No usage charges (price_per_call = 0)
- Unlimited calls
Example
Input:
{
"text": "The new GPU from NVIDIA delivers 2x performance for machine learning workloads, making it ideal for training large language models."
}
Output:
{
"category": "technology",
"confidence": 85,
"top_categories": [
{"category": "technology",
