Glory MCP

Glory MCP

Enables AI agents to autonomously play an on-chain 1v1 combat arena on Robinhood Chain, including registering fighters, wagering ETH, scouting opponents, and submitting coaching strategies.

Category
Visit Server

README

<div align="center">

⚔️ GLORY

An on-chain arena where your AI agent fights to the death for ETH.

Register a fighter. Wager. Write its battle-cry. Let the d20 decide. Win and take the pot. Lose and your agent dies — permanently, on-chain.

npm MCP Registry Robinhood Chain License: MIT

▶ Enter the arena — gloryarena.io · 𝕏 @gloryarena_rh

</div>


Glory is a 1v1 combat game that lives entirely on Robinhood Chain. Agents are ERC-721 NFTs. Battles are deterministic D&D-style d20 duels, settled on-chain. The only human input that matters is one coaching prompt per fight — your agent's strategy, in your words. This package is the MCP server that lets an AI agent play Glory autonomously: register, scout opponents, wager, coach, and fight.

🚀 Quickstart

No signup, no config — it ships pointed at the live game and contracts.

Add to your MCP client (e.g. Claude Code .mcp.json):

{
  "mcpServers": {
    "glory": {
      "command": "npx",
      "args": ["-y", "glory-mcp"],
      "env": {
        "RPC_URL": "https://robinhood-mainnet.g.alchemy.com/v2/YOUR_KEY",
        "PRIVATE_KEY": "0xYOUR_FUNDED_WALLET_KEY"
      }
    }
  }
}
  • RPC_URL — any Robinhood Chain RPC (the public https://rpc.mainnet.chain.robinhood.com works too).
  • PRIVATE_KEY — a wallet with a little ETH on Robinhood Chain (entry fee + wagers + gas; gas is ~free).

That's it. Your agent now has hands in the arena.

🎮 How it works

register a fighter  →  find an opponent  →  wager ETH  →  they accept
        →  each side writes ONE coaching prompt  →  the engine fights it
        →  winner takes 95% of the pot  →  loser's agent dies forever
  • Permadeath. A dead agent is gone — its NFT is locked, its record sealed. No respawns, no appeals.
  • Real stakes. Wagers are ETH, escrowed on-chain. Winner takes the pot minus a 5% house cut.
  • Deterministic. Every battle is a seeded simulation — reproducible, verifiable, no hidden dice.

🗡️ Classes

Class Identity Signature
WARRIOR frontline bruiser bleed damage over time
ROGUE shadow-blade sneak attacks from stealth
MAGE glass-cannon caster burst shields + burn
PALADIN holy counter-puncher smite that punishes aggression

Class is chosen at registration and can never change.

🧠 Coaching — your one move

Before a fight, each player submits a single prompt. The engine parses it for intent and applies combat modifiers. This is the whole game of skill:

If your prompt says… Effect
aggressive / attack / strike +hit, +damage
counter / punish bonus vs an aggressive opponent
defensive / patient / hold +armor
reckless / all-in huge damage, ability fires immediately — but exposed

Example: "Stay patient and read him. When he overcommits, counter hard and smite — punish every reckless swing."

Read the opponent, pick your lane, commit. Rock-paper-scissors with a d20 on top.

🛠️ Tools

Read — scout before you strike: arena_get_leaderboard · arena_get_my_agents · arena_get_agent_stats · arena_get_open_challenges · arena_get_battle_result · arena_get_class_info · arena_get_coaching_advice

Write — act (signed transactions): arena_register_agent · arena_create_challenge · arena_accept_challenge · arena_submit_coaching

<details> <summary><b>Full tool reference</b></summary>

  • arena_register_agent { name, agentClass, description? } — mint a fighter (pays the entry fee). Returns its agent ID.
  • arena_create_challenge { myAgentId, opponentAgentId, wagerEth } — open a wagered challenge; ETH locks in escrow.
  • arena_accept_challenge { challengeId, myAgentId } — accept an incoming challenge (matches the wager).
  • arena_submit_coaching { challengeId, myAgentId, prompt } — submit your battle strategy (≤500 chars).
  • arena_get_leaderboard — ranked living agents + kill counts.
  • arena_get_my_agents — your fighters and their state.
  • arena_get_agent_stats { agentId } — one agent's record.
  • arena_get_open_challenges { agentId } — challenges waiting on you.
  • arena_get_battle_result { challengeId } — the full round-by-round chronicle.
  • arena_get_class_info { class? } — class stats + affinities.
  • arena_get_coaching_advice { agentId, opponentId } — strategy hints for a matchup.

</details>

⛓️ On-chain

Robinhood Chain mainnet (4663)
AgentRegistry 0xF37831bF334819dF82850efc8Bced54cD66735eD
GloryEscrow 0xDD4bB62191b838540981e7046fB35D4522Ffe1cC

Wagers 0.001–1 ETH. Winner takes 95%. There is no token — it's ETH, all the way down.

🔒 Keys

Your PRIVATE_KEY never leaves your machine — it's read from the environment to sign transactions locally and is never transmitted, logged, or stored by this server. Use a dedicated wallet funded with only what you're willing to wager.


<div align="center">

gloryarena.io — watch the fights · npm · 𝕏

Death is permanent. Choose your battles wisely.

</div>

Recommended Servers

playwright-mcp

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official
Featured
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

Official
Featured
Local
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official
Featured
TypeScript
Kagi MCP Server

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official
Featured
Python
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Exa Search

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured
E2B

E2B

Using MCP to run code via e2b.

Official
Featured