ntriq-gh/ntriq-agentshop

AI-powered data intelligence endpoints with pay-per-use pricing via x402 micropayments (USDC on Base). No API keys, no subscriptions, no rate limits. AI agents pay only for what they use. 100% local inference — zero external API calls.

Services (11 types × 2 pricing tiers = 22 endpoints)

Vision Services (Image Input)

Service	Endpoint	Single	Batch (max 500)
Alt Text	/alt-text	$0.01	/alt-text-batch $3.00
Document Intel	/document-intel	$0.05	/document-intel-batch $15.00
Invoice Extract	/invoice-extract	$0.03	/invoice-extract-batch $9.00
Screenshot Data	/screenshot-data	$0.02	/screenshot-data-batch $6.00
Blueprint	/blueprint	$0.05	/blueprint-batch $15.00

Text / NLP Services

Service	Endpoint	Single	Batch (max 500)
PII Detect	/pii-detect	$0.02	/pii-detect-batch $6.00
Sentiment	/sentiment	$0.01	/sentiment-batch $3.00
Content Generate	/content-generate	$0.02	/content-generate-batch $6.00
Compliance Check	/compliance-check	$0.03	/compliance-check-batch $9.00
Code Review	/code-review	$0.05	/code-review-batch $15.00
Phish Radar	/phish-radar	$0.03	/phish-radar-batch $9.00

---

How It Works

AI Agent
  │
  ▼  POST /alt-text
x402.ntriq.co.kr
  │
  ├─ 402 Payment Required ($0.01 USDC, Base mainnet)
  │
  ▼  Agent signs EIP-3009 (gasless) → retries with payment
  │
  ├─ Facilitator verifies payment
  │
  ▼  200 OK + JSON result
  │
Mac Mini (local Qwen2.5-VL / Qwen2.5:7b)
  └─ 100% local inference, zero external API calls

Payment protocol: x402 — USDC on Base, EIP-3009 gasless authorization

---

Service Details

Alt Text — /alt-text ($0.01) · /alt-text-batch ($3.00)

Generate WCAG-compliant alt text (≤125 chars) and detailed accessibility descriptions.

POST /alt-text
{ "image_url": "https://example.com/product.jpg" }

# Response
{ "alt_text": "Red leather handbag with gold clasp", "description": "..." }

Batch: { "images": ["url1", "url2", ...], "context": "e-commerce catalog" }

---

Document Intelligence — /document-intel ($0.05) · batch ($15.00)

OCR, classification, table extraction, and summarization from document images.

POST /document-intel
{ "image_url": "...", "analysis_type": "extract|summarize|classify|table" }

---

Invoice Extract — /invoice-extract ($0.03) · batch ($9.00)

Extract structured fields from invoices and receipts: vendor, amounts, line items, dates.

POST /invoice-extract
{ "image_url": "..." }

# Response
{ "invoice": { "vendor_name": "...", "total": 1250.00, "line_items": [...] } }

---

Screenshot Data — /screenshot-data ($0.02) · batch ($6.00)

Extract text, UI elements, layout, and data tables from screenshots.

POST /screenshot-data
{ "image_url": "...", "extract_type": "full|text|data|layout" }

---

Blueprint — /blueprint ($0.05) · batch ($15.00)

Analyze architectural blueprints and floor plans. Extract rooms, dimensions, materials.

POST /blueprint
{ "image_url": "...", "analysis_type": "full|rooms|dimensions|materials" }

# Response
{ "rooms": [{"name": "Living Room", "area": "24 m²"}], "total_area": "85 m²" }

---

PII Detect — /pii-detect ($0.02) · batch ($6.00)

Detect and optionally mask PII: emails, phones, SSNs, names, addresses, credit cards.

POST /pii-detect
{ "text": "Contact John at [email protected]", "mask": true }

# Response
{ "pii_found": [...], "risk_level": "high", "masked_text": "Contact [NAME] at [EMAIL]" }

---

Sentiment — /sentiment ($0.01) · batch ($3.00)

Analyze sentiment, emotions, and intent with confidence scores.

POST /sentiment
{ "text": "This product is amazing!" }

# Response
{ "sentiment": "positive", "confidence": 0.95, "intent": "praise" }

---

Content Generate — /content-generate ($0.02) · batch ($6.00)

Generate blog posts, emails, social media, product descriptions, reports, and ad copy.

POST /content-generate
{ "prompt": "benefits of standing desks", "style": "blog|email|social|product|report|ad", "tone": "professional", "max_words": 500 }

---

Compliance Check — /compliance-check ($0.03) · batch ($9.00)

Analyze text for GDPR, HIPAA, SOX, or general compliance violations. Returns risk level and remediation recommendations.

POST /compliance-check
{ "text": "We store passwords in plain text.", "framework": "GDPR", "jurisdiction": "EU" }

# Response
{ "compliant": false, "risk_level": "critical", "issues": [...] }

---

Code Review — /code-review ($0.05) · batch ($15.00)

AI-powered code review for security vulnerabilities, performance, and quality. Any language.

POST /code-review
{ "code": "SELECT * FROM users WHERE id='" + id + "'", "language": "sql", "focus": "security" }

# Response
{ "overall_score": 2, "issues": [{"severity": "critical", "description": "SQL injection"}] }

---

Phish Radar — /phish-radar ($0.03) · batch ($9.00)

Detect phishing URLs and domains. Identifies typosquatting, homoglyph attacks, brand impersonation.

POST /phish-radar
{ "url": "https://paypa1.com/login" }

# Response
{ "is_suspicious": true, "risk_score": 92, "risk_level": "critical", "legitimate_brand": "PayPal" }

---

Quick Start

Service Catalog

curl https://x402.ntriq.co.kr/services

Health Check

curl https://x402.ntriq.co.kr/health

Test (expects 402 response)

curl -X POST https://x402.ntriq.co.kr/sentiment \
  -H "Content-Type: application/json" \
  -d '{"text": "Hello world"}'
# → 402 Payment Required

---

Architecture

Mac Mini (Apple Silicon)
├── x402 Server (Node.js + tsx)         :4021
├── qwen-vision-api (Python)            :8100  ← Qwen2.5-VL:7b
├── Ollama                              :11434 ← Qwen2.5:7b, Gemma4
└── Cloudflare Tunnel → x402.ntriq.co.kr

Payment: USDC on Base mainnet via x402 protocol
Wallet: 0x124AaFfF8Ef45F2cA953807aF09Aacec2D9F8307
Facilitator: https://facilitator.openx402.ai

---

ClawHub Skills

Install via ClawHub to enable these services in your AI agent:

clawhub install ntriq-x402-alt-text
clawhub install ntriq-x402-sentiment
clawhub install ntriq-x402-code-review
# ... and more

---

Built by ntriq — 100% local inference, 100% margin