About Pokerai API
Pokerai API’s MCP server gives poker students, coaches, and researchers access to presolved GTO strategy for 6-max No-Limit Hold’em, with optional on-demand postflop solving. Real-time assistance at real-money tables is prohibited.
Key features
- 01Preflop action frequencies
- 02Full preflop hand ranges
- 03Flop decision trees and nodes
- 04Optional flop, turn, and river solves
- 05Per-action expected values
Use cases
- 01Reviewing a preflop decision
- 02Studying a flop strategy node
- 03Comparing action values in a solve
MCP support
Exposes Pokerai API strategy lookups and, when enabled, real-time solver operations as tools for MCP clients. Requests use the operator’s API key and quota.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
npx -y @pokerai/mcpPOKERAI_API_KEYSecretYour Pokerai API key, from https://pokerai.bet/login. Calls spend your own quota.POKERAI_ENABLE_SOLVESet to 1 to expose the real-time solver tools, which spend the scarce solve quota. Off by default.Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.
Tools
5Saved tool list · updated 4d ago
This tool takes no parameters.
Details
From the project README@pokerai/mcp
MCP server for the Pokerai API — solver-grade GTO strategy for 6-max No-Limit Hold'em, exposed as tools your LLM agent (Claude, Cursor, Claude Code, …) can call.
Ask "what's GTO for AKs on the button facing a UTG open?" and the agent gets real presolved frequencies, not a guess.
Setup
You need a Pokerai API key — get one free at https://pokerai.bet/login.
The server runs over stdio via npx, so there's nothing to install globally.
Claude Desktop
Add to claude_desktop_config.json (Settings → Developer → Edit Config):
{
"mcpServers": {
"pokerai": {
"command": "npx",
"args": ["-y", "@pokerai/mcp"],
"env": { "POKERAI_API_KEY": "gto_your_key_here" }
}
}
}
Claude Code
claude mcp add pokerai -e POKERAI_API_KEY=gto_your_key_here -- npx -y @pokerai/mcp
Cursor
~/.cursor/mcp.json (or Settings → MCP), same shape as the Claude Desktop block above.
Tools
Presolved lookups (spend your presolved quota — 10/mo on the free tier):
| Tool | What it returns |
|---|---|
preflop_versions | selectable strategy chart sets (free) |
preflop_strategy | fold/call/raise frequencies for a hand vs an action line |
preflop_range | the whole 169-hand-type range for a position + line |
flop_tree | flop decision tree (nodes carry tokens); optional flop_version picks the dataset |
flop_node | strategy at one flop node (with or without a hero hand) |
Real-time solver (flop/turn/river) — off by default because it spends the scarce solve quota
(5/mo on the free tier). Enable with POKERAI_ENABLE_SOLVE=1:
| Tool | What it returns |
|---|---|
solve_schedule | trigger an on-demand solve (costs 1 solve quota) |
solve_tree | poll the solve + get its nodes |
solve_node | strategy at one solved node |
node_evs | per-hand, per-action EVs at a node |
Environment
| Var | Purpose |
|---|---|
POKERAI_API_KEY | required — your key from https://pokerai.bet/login |
POKERAI_ENABLE_SOLVE | set to 1 to expose the real-time solver tools |
POKERAI_API_BASE | override the API base (default https://pokerai.bet) |
Notes
- The server authenticates with your key and spends your quota — it holds no shared credential and has no abuse surface of its own.
- stdout is the MCP protocol channel; the server logs only to stderr.
About Pokerai API
Pokerai API provides solver-grade GTO strategy and poker computation APIs for training products, review tools, coaches, researchers, and AI agents.
Free tier: 10 presolved lookups + 5 real-time solves per month.
Real-time assistance (RTA) at real-money tables is prohibited.
- Website: https://pokerai.bet/
- Docs: https://pokerai.bet/docs
- API reference: https://pokerai.bet/reference
- OpenAPI: https://pokerai.bet/openapi.en.json
MIT licensed. Docs: https://pokerai.bet/docs · Reference: https://pokerai.bet/reference
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.