MCP server · Finance and crypto · official
Trading Research Agents MCP server
by BijectiveLabs · official
Ask your AI whether a trading idea survives its costs, whether a dataset is fit to test on, and what the records say.
This is a helper from Bijective Labs that lets your AI read the results of an automated trading research system. It is read-only, meaning it looks things up and explains them, and it never places an order. It is handy if you or your team run trading experiments and want plain answers about what passed, what was refused, and why.
What is an MCP server? The 30-second version
On its own, your AI can only chat with you. An MCP server is a small helper program that gives your AI one new skill, here a connection to the Trading Research Agents system. Once it is connected, your AI can ask that system questions and bring the answers back into your chat. You still ask in normal words, and the helper does the fetching.
What this MCP server does
You ask your AI something like whether a claimed edge clears the full round trip cost. Your AI passes the question to this helper, which reads the research records the system already wrote to disk. The helper sends back the figures and the decision, for example that the edge was 1.94 times the cost while the desk asked for 2 times. Your AI then explains that answer to you in the chat. Nothing is changed, and no order is sent anywhere.
What you can do with it
- Check whether a claimed edge clears the full round trip cost
- Ask whether a trade dataset is fit to test on
- See what a sealed pre-registered validation decided
- Read the kill ledger of ideas that were refused
- Pull up run reports from past cycles
- Ask why a specific trade was refused and which figure caused it
Try asking your AI
- “Does the edge on this BTC idea clear the round trip cost?”
- “Is this dataset fit to test on, or should I fix something first?”
- “What did the sealed validation decide, and what was the reason?”
- “Show me the kill ledger and the latest run report”
What it gives back to you
You get answers in plain text inside your chat: yes or no decisions, the numbers behind them, and short reasons. Refusals come with the figure that caused them, so you can see exactly where an idea died. Run reports and the kill ledger come back as lists you can read line by line. Nothing is rewritten or hidden.
Before you start
What you need
- Python 3.11 or newer
- uv, the tool that installs and runs it (the README shows the one-line install)
- A model key from a provider such as Groq, Google, Anthropic or OpenAI, set in your desk's .env file
Good to know
This server is read-only and no tool places an order, but it does read your desk's records and settings, so treat anything it shows you as private to your team.
Install it with your AI
Add Trading Research Agents MCP server to your AI, no technical skills needed
You don't install anything by hand. You copy one prompt, paste it into an AI that can work on your computer, and it checks, installs and connects the server for you, asking you when it needs something.
Sign in to get the install prompt
Members get a ready-made prompt that lets the Claude desktop app check Trading Research Agents MCP server, install it and connect it for them, step by step. You don't need any technical skills: you copy, paste and answer a few questions. Your connected AI can also find and install any of the 4,609 MCP servers here for you.
Who it's for
People who run or review automated trading research, such as quant researchers, trading desk analysts, and anyone checking whether a strategy idea is worth testing.
