Cut N Looks Developer Portal
Programmatic interfaces, Model Context Protocol (MCP) tools, and machine-readable resources for autonomous AI agents, LLM assistants, and partner booking integrations.
Agentic Architecture & Standards
Cut N Looks provides first-party support for AI agent discovery, structured data schemas, standard HTTP Markdown Content Negotiation (acceptmarkdown.com), and the Model Context Protocol (MCP).
Markdown Negotiation
Pass Accept: text/markdown to receive clean markdown on any page.
Streamable MCP Server
Full JSON-RPC 2.0 endpoint exposing booking and catalog tools at /api/mcp.
Machine-Readable Endpoints
| Resource | Endpoint | Format |
|---|---|---|
| LLM Context Index | /llms.txt | Text / Markdown |
| LLM Full Knowledge Base | /llms-full.txt | Text / Markdown |
| OpenAPI 3.1 Schema | /api/openapi.json | JSON (OpenAPI 3.1) |
| MCP Discovery Handshake | /.well-known/mcp | JSON |
| MCP Live Server | /api/mcp | JSON-RPC 2.0 |
| Agent Instructions | /.well-known/agent-instructions.md | Markdown |
| XML Sitemap | /sitemap.xml | XML |
Available MCP Tools
Returns the complete catalog of haircutting, balayage, skincare, and bridal couture services with exact pricing in INR and durations.
Returns master hair architects, senior barbers, bridal leads, and skincare aestheticians with background credentials and specialties.
Validates date availability and returns standard 90-minute private appointment slots between 10:00 AM and 9:00 PM IST.
Constructs a confirmed booking session and returns an instant WhatsApp concierge confirmation URI for the client.
Integration Examples
1. Fetching Markdown via Content Negotiation
curl -H "Accept: text/markdown" https://cut-n-look-web.vercel.app/services2. Calling MCP Tool (JSON-RPC 2.0)
curl -X POST https://cut-n-look-web.vercel.app/api/mcp \
-H "Content-Type: application/json" \
-d '{
"jsonrpc": "2.0",
"id": 1,
"method": "tools/call",
"params": {
"name": "list_services",
"arguments": { "category": "hair" }
}
}'All machine-readable endpoints comply with OWASP API standards, SSL TLS 1.3 encryption, and acceptmarkdown.com caching best practices.

