Intent Detection
Detect user intent from natural language messages for routing, support, and workflow automation.
/api/v1/ai/intent-detection
Loading RSFlowHub...
Sentiment analysis, content moderation, translation, PII redaction, ticket triage and more: 29 ready-to-use endpoints that return structured JSON, powered by OpenAI, Gemini, DeepSeek and Grok behind one API key. No prompt engineering, no parsing free text.
curl -X POST 'https://rsflowhub.com/api/v1/ai/sentiment-analysis' \
-H 'X-API-Key: YOUR_API_KEY' \
-H 'Content-Type: application/json' \
-d '{"text": "Checkout was fast, but support took 3 days to reply.", "aspects": ["checkout", "support"]}'
{
"success": true,
"request_id": "req_3f9a2c7e-5b1d-4e8a-9c2f-7d6e1a0b4c8d",
"data": {
"result": {
"sentiment": "mixed",
"confidence": 0.91,
"aspect_sentiments": [
{ "aspect": "checkout", "sentiment": "positive", "confidence": 0.94 },
{ "aspect": "support", "sentiment": "negative", "confidence": 0.9 }
]
}
},
"meta": { "version": "v1", "timestamp": "2026-10-05T09:30:00Z" }
}
Call any RSFlowHub endpoint with a standard REST request from your favorite language. You get structured JSON back, with automatic model fallback and a request ID on every call for easy debugging.
curl -X POST "https://rsflowhub.com/api/v1/ai/intent-detection" \
-H "Content-Type: application/json" \
-H "X-API-Key: rfh_your_api_key" \
-d '{
"text": "Please cancel my subscription and issue a refund to my credit card."
}'
import requests
url = "https://rsflowhub.com/api/v1/ai/intent-detection"
headers = {
"Content-Type": "application/json",
"X-API-Key": "rfh_your_api_key"
}
payload = {
"text": "Please cancel my subscription and issue a refund to my credit card."
}
response = requests.post(url, json=payload, headers=headers)
data = response.json()
result = data["data"]["result"]
print("Intent:", result["intent"], "Confidence:", result["confidence"])
const response = await fetch("https://rsflowhub.com/api/v1/ai/intent-detection", {
method: "POST",
headers: {
"Content-Type": "application/json",
"X-API-Key": "rfh_your_api_key"
},
body: JSON.stringify({
text: "Please cancel my subscription and issue a refund to my credit card."
})
});
const result = await response.json();
const { intent, confidence } = result.data.result;
console.log(intent, confidence);
<?php
$response = file_get_contents('https://rsflowhub.com/api/v1/ai/intent-detection', false, stream_context_create([
'http' => [
'method' => 'POST',
'header' => "Content-Type: application/json\r\n" .
"X-API-Key: rfh_your_api_key\r\n",
'content' => json_encode(['text' => 'Please cancel my subscription and issue a refund.'])
]
]));
$json = json_decode($response, true);
echo "Detected Intent: " . $json['data']['result']['intent'];
{
"success": true,
"request_id": "req_8b2d4f1a-6c3e-4a9b-8d7f-2e5c1b9a0f3d",
"data": {
"result": {
"intent": "cancel_subscription",
"confidence": 0.96,
"entities": { "refund_requested": true, "payment_method": "credit_card" }
}
},
"meta": { "version": "v1", "timestamp": "2026-10-05T09:30:00Z" }
}
Each endpoint is pre-prompted and tuned for one job and returns structured JSON. Pick what you need: intent classification, sentiment analysis, email replies, review moderation and more.
Detect user intent from natural language messages for routing, support, and workflow automation.
/api/v1/ai/intent-detection
Translate text between multiple languages using AI for high-quality contextual translations.
/api/v1/ai/translation
Standard conversational AI endpoint for interactive chat and assistant-like applications.
/api/v1/ai/chat
Generate intelligent, context-aware email replies based on the original email and optional analysis results. C...
/api/v1/ai/email-reply
Generate marketing copy, blogs, product text, and social content from short prompts.
/api/v1/ai/content-generation
Generate images from text prompts using advanced AI models.
/api/v1/ai/image-generation
Extract structured JSON data from unstructured text based on a schema.
/api/v1/ai/json-extractor
Analyze incoming emails using AI to extract structured insights such as category, intent, priority, sentiment,...
/api/v1/ai/email-analyze
Determine the most appropriate business action for an email using AI. The API analyzes the email and recommend...
/api/v1/ai/email-workflow
Analyze the emotional tone of text to identify positive, negative, neutral, or mixed sentiments.
/api/v1/ai/sentiment-analysis
Analyze text to determine if it is spam, promotional abuse, or malicious content.
/api/v1/ai/spam-detection
Condense long-form text, articles, or transcripts into configurable summaries and optionally extract structure...
/api/v1/ai/text-summarize
Automatically correct spelling, grammar, and punctuation while optionally adjusting tone and dialect.
/api/v1/ai/grammar-fix
Automatically detect and securely scrub Personally Identifiable Information (PII) from text.
/api/v1/ai/pii-redaction
Create and manage collections of FAQ items for semantic search matching.
/api/v1/ai/faq-collections
Match a search query semantically against an FAQ collection using a hybrid vector-similarity algorithm.
/api/v1/ai/faq-match
Analyze sentiment, category, key topics, emotions, and urgency of a customer review.
/api/v1/ai/review-analyze
Summarize a batch of customer reviews, extracting positive/negative themes, complaints, and suggestions.
/api/v1/ai/review-summary
Generate a professional, context-appropriate response to a customer review.
/api/v1/ai/review-reply
Analyze multiple reviews to generate actionable business recommendations and priorities.
/api/v1/ai/review-insights
Analyze sentiment, intent category, topics, reply requirements, recommended actions, and priority of a social...
/api/v1/ai/comment-analyze
Generate an engaging, context-aware reply to a social media comment matching your business context.
/api/v1/ai/comment-reply
Perform automated safety checks on social media comments to flag spam, abuse, harassment, and toxic behavior.
/api/v1/ai/comment-moderate
Analyze sentiment, priority, urgency, intent, department routing, and recommended actions of a customer suppor...
/api/v1/ai/ticket-analyze
Generate a professional, context-appropriate reply to a support ticket using only available details.
/api/v1/ai/ticket-reply
Determine the next automation workflow action, priority, and department route for a customer support ticket.
/api/v1/ai/ticket-workflow
Understand, normalize, and extract filters, exclusions, and sorting parameters from user natural-language sear...
/api/v1/ai/search-query-parse
Manage first-class AI knowledge bases, index FAQs, plain text, and PDF documents, and perform grounded vector...
/api/v1/knowledge-bases
Build customer-facing AI support chatbots with connected knowledge bases, client-managed actions, and automati...
/api/v1/chatbots
Stop fighting raw LLM formatting inconsistencies, unexpected upstream vendor downtime, and separate billing accounts.
Every endpoint returns the same envelope: { success, request_id, data: { result }, meta }. Results are validated before they are returned, so your code works with fields, not free text.
Requests run on models from OpenAI, Google Gemini, DeepSeek and xAI Grok. We pick and tune the model per task, so you never write or maintain prompts yourself.
A circuit breaker tracks provider health. When a provider returns errors or rate limits, the request is retried on a backup model, so a single upstream outage does not take your feature down.
Skip separate accounts, keys and invoices for every AI vendor. One RSFlowHub key, one credit balance, billed in USD ($) or INR (₹).
Upload documents or FAQ items, then search them semantically or ask questions with cited answers. Ideal for support bots and internal help centers.
RSFlowHub never uses your requests to train models. We log usage metadata for billing, and identical requests may be cached for up to 24 hours so repeats cost less.
See why developers choose RSFlowHub over managing direct AI provider keys or generic unvalidated proxies.
| Platform Capability | RSFlowHub | Direct Provider APIs | Generic LLM Proxies |
|---|---|---|---|
| Master API Authentication | One Key for Every Endpoint | One Key per Provider | Single Proxy Key |
| JSON Response Reliability | Validated, Structured JSON | Raw Text / Prompt Drift | Raw Text Forwarding |
| Automatic Provider Failover | 3-State Circuit Breaker | Manual Client-Side Retries | Basic Fallback |
| Pre-Built Task Endpoints | 29 Ready-Made Endpoints | 0 (Prompt Eng. Required) | 0 (Pass-through only) |
| Unified Wallet & Multi-Currency | Single Balance (USD $ / INR ₹) | 5+ Credit Cards / Billing Rows | USD Credit Cards Only |
Native Vector RAG (pgvector) |
Built-in Knowledge Bases | Requires External Vector DB | No Vector Storage |
| Training on Your Data | Never Used for Training | Vendor Specific Retention | Varies by Proxy Host |
Start making production AI API calls in under 2 minutes.
Sign up with email, Google or GitHub. The Free plan includes 100 credits every month, and verifying your phone adds 1,000 bonus credits.
Call any endpoint from cURL, PHP, Node.js, Python or any language that can make an HTTPS request, using your single X-API-Key.
Most calls return in 1 to 3 seconds with a consistent JSON envelope, so you can store, display or act on the result immediately.
Pay only for what you consume. Atomic credit pre-authorization holds ensure zero hidden fees or surprise invoices.
testing and personal prototypes
indie developers shipping MVPs
production apps with steady traffic
high-volume client and agency workloads
Everything you need to know about API keys, dynamic routing latency, credit billing, and data privacy.
RSFlowHub is a developer platform of ready-made AI APIs. Instead of writing prompts and parsing free text, you call an endpoint such as sentiment analysis, content moderation, translation or ticket triage and receive structured JSON. 29 endpoints are available with one API key.
No. One RSFlowHub API key covers every endpoint. Provider accounts, model selection, prompts, retries and token costs are handled on our side and billed to you as simple credits.
Most requests complete in 1 to 3 seconds, depending on input length and the AI model. Identical repeat requests are served from cache and return faster. We recommend a client timeout of 15 to 30 seconds for generation endpoints.
A circuit breaker monitors each provider. When a provider returns errors or rate limits, requests are retried on a configured backup model, and you are not charged for failed calls.
Every successful call returns { "success": true, "request_id": "req_...", "data": { "result": { ... } }, "meta": { "version": "v1", "timestamp": "..." } }. Errors use the same envelope with an error code and message, so one parser handles every endpoint.
Each call costs credits: simple calls start at 1 base credit, and generation endpoints add credits for longer inputs and outputs. The Free plan includes 100 credits per month plus 1,000 free requests on utility APIs (intent, sentiment, PII redaction and spam detection). Paid plans start at $6/month. Extra credit packs start at $2 and never expire.
RSFlowHub does not use your requests or outputs to train AI models. Request text is sent to the selected AI provider only to produce the result. We log usage metadata such as character counts and credits for billing, and identical requests may be cached for up to 24 hours so repeat calls cost less.
Any language that can make an HTTPS POST request. The documentation includes copy-paste examples for cURL, PHP, Node.js and Python, and the endpoints work the same from Laravel, Express, Next.js, Django or a mobile backend.
Add sentiment analysis, moderation, translation and support automation to your product today. Free plan with 100 credits/month + 1,000 bonus credits when you verify your phone.
🐞 Found an issue? Report it and our team replies within 1 business day.
💡 Need a custom endpoint? Tell us what you need and we will scope it with you.