Documentation Overview
Everything you need to integrate OnSaleNow.ai with your AI applications
What is MCP?
The Model Context Protocol (MCP) is a standardized way for LLMs to discover and invoke external tools. Instead of hard-coding integrations for each AI model, MCP provides a universal interface: LLMs can query available tools, read their schemas, and call them with structured arguments.
OnSaleNow.ai exposes deal data through MCP-compatible servers. Any LLM runtime that supports MCP (Claude Desktop, LangChain, custom agents, etc.) can search, filter, and retrieve deals without custom code.
OnSaleNow.ai MCP Tools Overview
findDeals
Search for deals by query, category, price range, merchant, and freshness. Returns paginated results sorted by discount, price, or recency.
getDealById
Retrieve detailed information about a specific deal by its unique ID, including full description, images, and merchant metadata.
Authentication
All API requests require a Bearer token in the Authorization header:
Authorization: Bearer YOUR_API_KEYYou'll receive an API key when you sign up. Tokens can be scoped and rotated from your dashboard.
Filter Parameters
The findDeals tool supports rich filtering:
- query: Text search (e.g., "wireless headphones")
- category: Filter by category (e.g., "Electronics", "Home & Garden")
- price_range: Min/max price bounds (e.g., min=50, max=200)
- merchant: Whitelist or blacklist specific merchants
- min_discount: Minimum discount percentage (e.g., 25 for 25% off)
- updated_since: ISO timestamp for real-time feeds (e.g., "2025-10-01T00:00:00Z")
- sort_by: Sort results by "discount", "price", or "updated_at"
Pagination & Rate Limits
Results are paginated with a default of 10 items per page. Specify max_results (up to 100) and use offset for pagination.
Rate limits depend on your plan. Exceed them and you'll receive a 429 Too Many Requests response. Monitor usage in your dashboard or set hard caps to avoid overages.