14 articles

AI assistants & MCP

Using ChatGPT, Claude, Claude Code and Codex to research, build and backtest trading strategies over MCP, safely.

AI Trading Copilot: Plain English to a Backtested Strategy

How an AI trading copilot turns plain English into a backtested strategy in Trigr Studio, and the checks to run before you trust its result.

7 min read

Best MCP Server for Trading: Execution vs Validation

The best MCP server for trading depends on the job. Compare Coinbase, OKX, Alpaca, CCXT and Hyperliquid MCPs with Trigr's validation-only server.

6 min read

Claude for Trading: A Research Workflow to a Paper Agent

A step-by-step Claude for trading workflow: pre-register a hypothesis, search the catalog, backtest, run experiments, freeze a winner and paper-test it.

6 min read

Crypto MCP Servers: A Research Stack for Claude and Codex

Crypto MCP servers from Nansen, Dune, CoinGecko and others, how to run them in Claude Code, Codex or Cursor, and how to end the research in a Trigr backtest.

7 min read

Iterate on a Trading Strategy With AI Using Experiments

How to iterate on a trading strategy with AI: labelled experiments inside one strategy's version history, not dozens of near-duplicate strategies.

7 min read

LLM Trading Strategy: Can You Trust What the Model Wrote?

Can you trust an LLM trading strategy? The guardrails that make AI-written strategies checkable: catalog validation, recipe hashes, approvals and trade logs.

6 min read

TradingView MCP: Import a Pine Strategy Into Trigr

Use the TradingView MCP and Trigr's MCP to port a Pine Script strategy into Trigr, backtest it with costs, iterate in experiments and create a paper agent.

9 min read

Why Most AI Trading Bots Fail, and What Honest Looks Like

Why most AI trading bots fail: hallucinated edges, look-ahead bias, missing costs and unlimited authority. What an honest AI trading setup looks like.

6 min read

AI Trading Agent Safety: Approvals and Hard Limits on Trigr

AI trading agent safety on Trigr: exact-argument approvals, manual vs autonomous mode, and hard limits that stop assistants from live trading or withdrawing.

6 min read

AI Trading Strategy Prompts: Recipes to Build and Test

Copy-ready AI trading strategy prompts for ChatGPT, Claude and Codex: build, backtest net of costs, improve, and research the marketplace without overfitting.

6 min read

Claude Code Trading: Connect Claude Code and Codex to Trigr

Claude Code trading setup: connect Claude Code and Codex to Trigr's MCP server, authorize in your browser, and backtest crypto strategies from the terminal.

6 min read

MCP vs API for Trading: Giving an LLM Market Access

MCP vs API for trading: how Model Context Protocol, direct function calling and rules-based trading bots differ in setup, control and safety for LLMs.

6 min read

Trading MCP Server for ChatGPT and Claude: Trigr Explained

How Trigr's trading MCP server lets ChatGPT, Claude, Claude Code and Codex research, build and backtest Hyperliquid strategies, with approvals and hard limits.

9 min read

Use ChatGPT to Build and Backtest a Crypto Strategy

A step-by-step guide to using ChatGPT to build and backtest a crypto strategy on real point-in-time data with Trigr's MCP server, gross and net of costs.

6 min read

Put the idea to an honest test.

Describe a strategy in plain English or from your own AI assistant, backtest it on point-in-time data, and forward-test it on paper before any real money is involved.