Reference

MCP server

The Model Context Protocol server exposes the platform as eight tools an AI assistant can call directly. Each tool is a thin wrapper over the REST surface and inherits the same metering, provenance, and honesty gate.

Rolling out. The MCP endpoint ships with the Ultra (Enterprise) MCP release. The tools and contract below are final; the connection URL becomes live at that point. The REST API is available today.

Transport

Streamable HTTP at https://patternsight-platform.fly.dev/mcp. Compatible with Claude, direct MCP clients, and the OpenAI Responses API.

Auth

Authorization: Bearer with your Ultra API key. ChatGPT connects over OAuth instead — a separate, later release.

Entitlement

Data tools work with any platform key; generate_picks and backtest require the Ultra (Enterprise) tier.

Tools

list_gamesmeter: api_requests

List available lottery games and whether picks/backtest are supported for each.

get_gamemeter: api_requests

Get one game's number configuration (pool, range, special ball).

latest_resultsmeter: api_requests

Most recent draw result(s) for a game. Past results only — descriptive.

frequencymeter: api_requests

Number-frequency analysis with a chi-squared residual.

Describes past draws — not a prediction of future draws.

gapsmeter: api_requests

Gap analysis: draws since each number last appeared, with mean/median/max gap.

A long gap does not make a number "due" — draws are independent.

analyze_gamemeter: api_requests

Composed descriptive analysis — frequency, gaps, and draw uniformity from precomputed digests.

generate_picksmeter: agentic_analysesUltra (Enterprise) tier

Generate number picks using the swarm's mathematical strategies. Each result carries a provenance pack_hash and the disclaimer.

Picks are mathematical selections, not predictions; they do not improve your chances of winning.

backtestmeter: agentic_analysesUltra (Enterprise) tier

Walk-forward backtest of a strategy against historical draws, with no lookahead.

edge_vs_random clusters near 0 — that means no real edge. A small non-zero value is sample noise, not a winning signal: no strategy beats random on fair lottery draws.