Research by conversation
Ask the agent to investigate markets, companies, filings, strategies, and your connected account context.
BotSpot trading guides
Build, test, connect, and operate trading strategies
Follow focused guides for choosing a workflow, turning rules into inspectable code, testing historical evidence, controlling deployment risk, and connecting ChatGPT, Claude, or Codex to supported broker workflows.
BotSpot capabilities
Place approved direct trades
When direct trading is enabled, request a one-time trade in plain English, inspect the order, and approve it before submission.
Build complete algorithms
Create, revise, backtest, connect, and operate Lumibot strategies without assembling the full application stack yourself.
Use the AI client you prefer
Work in BotSpot or connect through ChatGPT, Claude, Cursor, Codex, and other compatible MCP clients.
Choose a guide by the decision you need to make
Start with platform selection when choosing tools. Use build guides to formalize strategy rules, backtesting guides to inspect evidence, and deployment guides to validate broker and operational controls.
Treat every backtest as evidence with limits
Historical results depend on data, timing, costs, fills, parameter choices, and software behavior. Use the linked guides to inspect those assumptions instead of treating one performance number as proof.
Keep broker access and high-risk actions controlled
Broker support, assets, account permissions, paper or live modes, and authentication methods vary. Connect credentials through BotSpot, keep secrets out of prompts, and review high-risk actions through the required approval flow.
Explore the right path
Automated trading buyer guide
Choose the workflow, not just the backtest tool
Choose among an AI trading workspace, quantitative platform, broker API, or open-source framework by comparing the complete workflow.
Practical AI trading bot guide
Build an AI trading bot from testable rules, not a vague prompt
Turn a trading idea into explicit rules, inspect AI-generated strategy code, backtest it, test it with paper trading, and prepare it for controlled deployment.
Plain-English algorithmic trading guide
Build an algorithmic trading workflow without writing every line yourself
Learn how to describe trading rules in plain English, generate inspectable strategy code with AI, backtest it, and validate it without writing every line yourself.
AI strategy backtesting guide
Backtest an AI trading strategy without mistaking simulation for proof
Build a defensible AI strategy backtest by freezing the hypothesis, checking data timing, documenting execution assumptions, and testing robustness.
Trading deployment guide
Paper trading vs live trading: know what each stage can prove
Compare paper and live algorithmic trading, learn what simulations can validate, and use a practical checklist before putting capital at risk.
Backtesting software buyer guide
Best backtesting software depends on your strategy workflow
Compare backtesting software for AI-assisted, no-code, chart-based, code-first, and MQL5 workflows using documented features and limitations.
Backtest performance metrics
Backtesting metrics explained without magic numbers
Learn how CAGR, volatility, Sharpe, Sortino, drawdown, win rate, profit factor, expectancy, and turnover describe a trading backtest.
Trading strategy validation guide
Avoid overfitting a trading strategy before strong backtest results mislead you
Reduce backtest overfitting by logging every trial, separating strategy selection from evaluation, preserving unseen data, and testing parameter robustness.
Trading strategy validation guide
Use walk-forward testing to measure repeated out-of-sample behavior
Learn how walk-forward testing uses ordered training and test windows to evaluate trading strategies while reducing leakage and overfitting risk.
Backtest data-integrity guide
Stop future information from leaking into your backtest
Learn how point-in-time data, historical universes, filing timestamps, corporate actions, and indicator alignment prevent false backtest results.
Backtesting execution-cost guide
Model slippage, fees, and market impact without making a backtest look safer than it is
Learn how to model commissions, bid-ask spreads, slippage, fill timing, liquidity, partial fills, and market impact without inventing universal assumptions.
Automated trading risk-control guide
Build algorithmic trading risk controls around every order and failure state
Design algorithmic trading controls for position sizing, exposure, leverage, order limits, drawdowns, monitoring, broker constraints, and emergency stops.
ChatGPT broker connection guide
Connect ChatGPT to your broker through BotSpot
Connect ChatGPT to BotSpot through OAuth MCP, verify your supported broker and account mode, and keep every trading action behind explicit approval.
Claude and broker connection guide
Connect Claude to your broker through BotSpot
Connect Claude to BotSpot through OAuth, use supported broker context, understand trade approvals, and keep brokerage credentials out of AI prompts.
Codex MCP broker guide
Connect Codex to your broker through BotSpot
Configure BotSpot MCP in Codex, keep broker credentials inside BotSpot, inspect trading context, and stage broker orders behind explicit approval.
Sources
- BotSpot MCP for Agents
- BotSpot broker connection manifest
- BotSpot Terms of Service
- Lumibot documentation
Sources verified 2026-08-17.