เซิร์ฟเวอร์ MCP Tradefloor

Overview

tradefloor is a simulated stock market, with a limit order book and an economy, for testing trading strategies and AI agents. A market is computed from its inputs, so the same inputs give the same market again, and an agent's own orders move the prices it trades at.

Start with a task

New to tradefloor

Run your first market. Install, run a five-company market for 20 days, score an agent and read the result, in five minutes.

Building an agent

Write an agent. The act(obs) method, the orders it can return, sizing to a target weight and what the agent can see.

Evaluating strategies

Compare strategies. Against buy-and-hold on one market, then across many seeds, and by what each one's trades cost.

Testing counterfactuals

Fork a market. Copy a running market, change one input in one copy, and measure what the change did.

More workflows

Run a scenario: score an agent with and without a shipped scenario on the same seeds.
Test an LLM agent: PydanticAI, the OpenAI Agents SDK, LangGraph or a plain function, with its answers recorded.
Record and replay: run a recorded LLM agent again without calling the model.
Reproduce or publish an experiment: what fixes a market, saving a run as a manifest, and what to report (Citing tradefloor).
Train an RL policy: a Gymnasium environment whose reward includes the cost of the policy's own trading.
Connect through MCP: a local server that lets a model build markets, score strategy specs and explain price moves.

Concepts

UniversePresetScenarioSeed

Engine the market: order book, companies, economy

observation →← orders

Agent act(obs)

Run the engine stepped through trading days

Score evaluate, rank

Checkpoint, then fork two futures from one past

Manifest, then reproduce inputs, version and orders, saved

EngineOne market. tf.Engine(seed=..., universe=..., model=...) runs the order book, the companies' fair values and the economy, and returns Arrow tables of what happened. Engine

UniverseThe companies a market trades, in order. The order is part of the market. tf.Universe.random(n, seed=...) makes made-up companies, and Universe.from_edgar reads real ones. Universe

PresetA named, frozen set of model coefficients, such as pt-v20, the default. A shipped preset never changes. ModelParams

SeedThe integer that fixes every random draw the engine takes. Universe.random takes a seed of its own, which picks the companies. Seed

ScenarioA named set of changes to the economy or the market on given days, such as a rate shock, with the assumptions behind them stated. Run a scenario

AgentAny object with an act(obs) method. Strategies, LLM agents and RL policies all reach the market this way. Write an agent

Observation and actionAt each decision step the agent receives an observation (prices, its portfolio, the clock) and returns an action: orders, as shares per ticker, or None. What act returns

RunAn engine stepped through trading days. A run with an agent in it depends on the agent's orders too, because they fill against the book and move prices.

Score tf.evaluate runs each agent on its own copy of one market and returns a Scorecard. tf.rank repeats that over many seeds and counts which agent did better. Compare strategies

ForkA copy of a running market. Change one input in one copy and any later difference comes from that change. Fork a market

ManifestA RunManifest holds the version, preset, seed, universe, scenario and order log of a run, and rebuilds the market from them. Reproducibility

In code, the smallest complete run is a universe, an agent and a score:

import tradefloor as tf

universe = tf.Universe.random(5, seed=11)      # the companies, in order
agent = tf.baselines.BuyAndHold()              # act(obs) returns orders

# One engine per agent, on the default preset, seed 42, for 20 trading days
scores = tf.evaluate({"hold": agent}, seed=42, universe=universe, days=20)
print(scores["hold"])

Scorecard('hold', pnl=23,765, return=+2.38%, trades=5, impact=+0.43bps, sharpe=+1.63, vol=19.3%, in_market=100%, exposure=0.94x)

Reference

The recommended API is the short list of classes and functions most code needs. The complete index lists every export with its tier. The model is documented in How prices are made, its coefficients in Model parameters, and what has been checked against real markets in How it is measured. Support policy says which releases get fixes and what a patch release may change.

The hosted app at app.tradefloor.dev is documented separately under Hosted (preview).

Tradefloor MCP

เอกสาร