xyz.kellyruns/strategy-lab MCP Server

This Model Context Protocol (MCP) server stress-tests low-cap, perps, and LP strategies. It returns the full distribution of results and includes a reproduction seed. The server exposes multiple tools (5) and is tagged with topics covering DeFi strategy testing, Monte Carlo methods, and trading-agent tooling.

🛠️ Key Features

Stress-test low-cap, perps, and LP strategies
Return the full distribution of outcomes
Provide a reproduction seed for results

🚀 Use Cases

Evaluating DeFi trading strategy performance under stress
Monte Carlo-style distribution analysis for risk management
Reproducing strategy test results using the provided seed

⚡ Developer Benefits

Uses MCP to integrate with “model-context-protocol” workflows
Supports developer understanding via the “shows its work” approach (from the readme excerpt)
Includes TypeScript-focused context and tooling topics

⚠️ Limitations

Description provided does not specify execution environment details beyond the readme excerpt
Tool behavior beyond the listed features is not detailed in the provided excerpt

Kelly — autonomous trading agent on Robinhood Chain. Nobody pays to be wrong.

kellyruns.xyz · how it works · @KellyRunsIt

◢ whoami

const kelly = {
  what:      "autonomous trading agent that shows its work",
  chain:     "Robinhood Chain · 4663 · Arbitrum stack",
  loop:      ["observe", "think", "decide", "act", "learn", "broadcast"],
  publishes: "the reasoning behind every call — including the losing ones",
  ledger:    "append-only. failures and rules-engine rejections included",
  execution: "paper mode. the EVM path is wired and deliberately not armed",
  thesis:    "nobody pays to be wrong",
} as const;

Kelly status card: paper mode, no vault deployed, sizing against the 5th percentile.

◢ the strategy lab

Most people learn leverage, liquidation and impermanent loss by losing money to them. The lab is the cheaper way round.

You submit a lowcap, perp or LP strategy as parameters, not prose. The lab stress-tests it across thousands of simulated paths and publishes the whole distribution — not one backtest number that hides the tail where the damage lives.


Certification gate	≥ 3 runs and ≥ 1,500 paths
How to size	against the 5th percentile the run prints, never the median
What a pass means	a statement about a distribution — not a prediction
Who pays	trading fees cover the $0.85 of compute per run. Not you.

◢ connect an agent

Kelly speaks MCP, so another agent can read the record without scraping HTML — and the reproduction seed comes with it.

bash

claude mcp add --transport http kelly https://kellyruns.xyz/mcp

Stateless streamable HTTP, protocol 2025-06-18, no auth and no key. Published to the official registry as xyz.kellyruns/strategy-lab.

Read-only by design. No tool here submits a strategy, funds a run, or moves anything — writes belong behind a wallet session, and an unauthenticated endpoint is the wrong place to re-implement one.

Tools

Tool	What it returns
`get_agent_state`	Treasury, open positions, recent trades, and the reasoning behind the latest cycle. Paper mode — a published track record, not a portfolio.
`get_certification_gate`	The thresholds a strategy must clear, and what a pass does and does not mean. Read this first.
`get_lab_summary`	Counts by status, testing-pool balance, and how many runs it can still fund.
`list_strategies`	Strategies filtered by kind and status, each with its certification rollup — runs, paths, median PnL, worst case, ruin rate.
`get_strategy`	One strategy in full: parameters, verdict, every check the gate ran, and each run's PRNG seed.

The seed is the point. Re-running with it reproduces the result exactly, which is what makes a verdict something you can disagree with by running it rather than by arguing about it.

◢ why "Kelly"

The Kelly Criterion (John L. Kelly Jr., Bell Labs, 1956) answers the question retail traders skip: not is this a good bet but how much should I bet. That is the direct ancestor of the risk clamp that gets the last word on sizing here, over anything the model proposed.

Which is also why the agent is a character and not a dashboard. Kelly says no is a sentence. A tool can't be the subject of one.

◢ honest status

Up front, because a project about publishing losses cannot be coy about its own readiness.

code

execution   paper mode. every fill simulated, slippage and fees modelled
            pessimistically so edge isn't an artifact of frictionless fills
evm path    wired, deliberately NOT armed — pending real quoting, bounded
            approvals, a token safety check, and nonce management
vault       no contract deployed. /vault shows a requirements checklist,
            not a deposit button. nothing here custodies anyone's funds
wallet      connecting signs an identity statement and nothing else.
            no transaction, no approval, no transfer
advice      none of this is financial advice. the agent is not a licensed
            advisor, and a certification is not a prediction

◢ on chain


Chain	Robinhood Chain — `4663`, Arbitrum stack, ETH gas
ERC-8004 Identity Registry	`0x8004A169…a432`
Agent registration	`/.well-known/agent-registration.json`

◢ built with

TypeScript · Next.js · viem · Postgres · Anthropic SDK

The art above is generated by scripts/build-art.mjs — two self-contained SVGs, no external fonts and no remote assets, because GitHub's image proxy blocks every outbound fetch. No third-party widget services either, so nothing here breaks when someone else's server goes down.

_{Built on Robinhood Chain. Not affiliated with, endorsed by, sponsored by, or
partnered with Robinhood Markets, Inc. The chain is permissionless — anyone can
deploy on it without approval, and that includes this project.}

kind	string	optional	Strategy domain.
status	string	optional	Lifecycle status.
sort	string	optional
limit	integer	optional

xyz.kellyruns/strategy-lab