MCP

Definition

MCP (Model Context Protocol) is an open protocol, originally developed by Anthropic, that standardizes how AI models interact with external tools and data sources. Instead of building custom integrations for each AI coding agent (Claude Code, Cursor, Codex, etc.), tools implement the MCP specification once and work across all MCP-compatible agents.

Tierux is MCP-first — every operation available in the dashboard (creating products, mapping entitlements, building paywalls, running test purchases) is also available as an MCP tool. Your AI agent can set up the entire billing stack by calling these tools, previewing the results, and committing when you confirm.

Examples

• A developer connects the Tierux MCP server to Claude Code with one JSON config file. They then type "Set up a $9.99/month Pro subscription" — the agent handles the entire setup.
• create_play_product accepts { productId, billingPeriod, price, currency, appId? }; create_product_mapping accepts { productId, entitlementId, productType, appId? }, where productType is subscription or one_time.
• create_paywall accepts a fully authored paywall object: id, theme, accent, badge, headline, subtitle, features, packages, recommended, both CTA labels, localizations, and optional URLs / appId. There is no template shortcut. Each write returns a preview; you confirm before anything is persisted.
• MCP tools also support read operations — the agent can list existing products, check paywall configurations, and verify entitlements for debugging.