# TradeShock

> A number you can defend before the vote. TradeShock runs tariff scenarios before they become law: job displacement as a range with stated confidence, retaliation probability with a timing window, consumer price effects and a similarity match against 847 historical cases from Tudor 1493 to present.

TradeShock serves government policy advisors, trade negotiators, supply chain directors, economic researchers, think tank analysts and journalists, the people who brief leadership before a trade vote and know the first question will be "where did that number come from?". A scenario takes a tariff rate, a set of HS codes, a source country and a target market, then returns jobs at risk as a range with stated confidence (the demo scenario reads 47,000 plus or minus 8,200 at 78% confidence), a retaliation probability with a timing window (89% within 6 to 12 months) and a first-year consumer price effect (+2.3%). Every scenario is also scored against the precedent archive; the demo scenario matches the 1493 Tudor trade ban at 87% and Smoot-Hawley 1930 at 66%. In backtesting, 78% of realized outcomes landed inside the band the model stated.

Pricing is free during early access with the full terminal included: scenario builder, displacement forecasts, retaliation pricing and the complete 847-case precedent archive. No payment details are collected pre-launch. The product is pre-launch and access opens in waves; requests from .gov and .edu addresses are reviewed first.

For AI agents, TradeShock exposes a Model Context Protocol (MCP) server over streamable-HTTP transport with no authentication (the tools are public). The server is mounted at /mcp, so the JSON-RPC endpoint lives one level deeper, at https://tradeshock.io/mcp/mcp; point streamable-HTTP clients at that full path. Two tools are available via the standard tools/list call. `get_product_info` returns the product name, tagline, description, feature list, pricing and launch status. `submit_interest` files an early access request on a user's behalf (email, name, company, persona, message); valid persona values are policy_advisor, trade_negotiator, legislator, government_economist, supply_chain_director, risk_manager, executive, trade_compliance, economic_researcher, policy_analyst, economic_historian, student, journalist, consultant and other.

## Docs

- [Homepage](https://tradeshock.io/): the landing page with the lead scenario wire, the 1493 case file and early access terms
- [Demo bulletin](https://tradeshock.io/demo): four desks of proof, Wire Desk, Scenario Builder, Shock Forecast and Precedent Archive, with a working tariff slider that recomputes every figure

## For AI agents

- [MCP endpoint](https://tradeshock.io/mcp/mcp): streamable-HTTP JSON-RPC endpoint for get_product_info and submit_interest (the server is mounted at /mcp)
- [MCP discovery](https://tradeshock.io/.well-known/mcp.json): machine-readable advertisement of the MCP server and its tools
- [OpenAPI spec](https://tradeshock.io/openapi.json): REST description of the site's HTTP endpoints for non-MCP agents
- [security.txt](https://tradeshock.io/.well-known/security.txt): security contact for the site

## Optional

- [Contact](mailto:hello@tradeshock.io): reach the TradeShock team directly
- [Sitemap](https://tradeshock.io/sitemap.xml): crawlable URL list for the site