Can I insert new test data dynamically tracking specific limits?

Yes. Utilizing the `insert_dataset_row` method, you can effortlessly inject exact JSON tracking payload mapping strings directly inside the text corpus evaluating the final results.

Does it pull out original Prompt definitions stored securely?

Certainly. The `get_prompt` command isolates and returns perfectly version-controlled bounding parameters slicing literal templates natively hosted under the Braintrust database.

How deeply can it inspect test regressions or scoring limits?

Using the robust `list_experiments` call, you can branch full arrays separating LLM version behaviors over massive iterations tracking the performance anomalies accurately.

Braintrust MCP Server for VS Code Copilot 10 tools — connect in under 2 minutes

Name: Braintrust
Availability: InStock
Author: Vinkius

code.visualstudio.com/docs/copilot/chat/mcp-servers

Built by Vinkius GDPR 10 Tools IDE

GitHub Copilot in VS Code is the most widely adopted AI coding assistant, embedded directly into the world's most popular code editor. With MCP support in Agent mode, Copilot can access external data and APIs to generate context-aware code grounded in real-time information.

Get MCP Server for AI Agents

ASK AI ABOUT THIS MCP SERVER

Open in ChatGPT Open in Claude Open in Perplexity

Vinkius supports streamable HTTP and SSE.

RecommendedModern Approach — Zero Configuration

Vinkius Desktop App

The modern way to manage MCP Servers — no config files, no terminal commands. Install Braintrust and 2,500+ MCP Servers from a single visual interface.

Download Free Open SourceNo signup required

Classic Setup·json

{
  "mcpServers": {
    "braintrust": {
      "url": "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
    }
  }
}

Fully ManagedVinkius Servers

60%Token savings

High SecurityEnterprise-grade

IAMAccess control

EU AI ActCompliant

DLPData protection

V8 IsolateSandboxed

Ed25519Audit chain

<40msKill switch

Stream every event to Splunk, Datadog, or your own webhook in real-time

* Every MCP server runs on Vinkius-managed infrastructure inside AWS - a purpose-built runtime with per-request V8 isolates, Ed25519 signed audit chains, and sub-40ms cold starts optimized for native MCP execution. See our infrastructure

About Braintrust MCP Server

Connect your Braintrust AI observation platform to any agent and maintain intense logic evaluation capabilities directly over conversation.

GitHub Copilot Agent mode brings Braintrust data directly into your VS Code workflow. With a project-scoped config, the entire team shares access to 10 tools. Copilot queries live data, generates typed code, and writes tests from actual API responses, all without leaving the editor.

What you can do

Project Analytics — Retrieve logic banks and branch isolated AI test sets
Experiments — Create real trace regression tests appending unique LLM scoring iterations
Datasets — Query accurate Ground Truth sets and insert new prompt templates mapping your system accuracy
Prompt Versioning — Grab perfectly frozen semantic prompts without editing core code boundaries

The Braintrust MCP Server exposes 10 tools through the Vinkius. Connect it to VS Code Copilot in under two minutes — no API keys to rotate, no infrastructure to provision, no vendor lock-in. Your configuration, your data, your control.

How to Connect Braintrust to VS Code Copilot via MCP

Follow these steps to integrate the Braintrust MCP Server with VS Code Copilot.

Create MCP config

Create a .vscode/mcp.json file in your project root

Add the server config

Paste the JSON configuration above

Enable Agent mode

Open GitHub Copilot Chat and switch to Agent mode using the dropdown

Start using Braintrust

Ask Copilot: "Using Braintrust, help me...". 10 tools available

Why Use VS Code Copilot with the Braintrust MCP Server

GitHub Copilot for Visual Studio Code provides unique advantages when paired with Braintrust through the Model Context Protocol.

VS Code is used by over 70% of developers. adding MCP tools to Copilot means your team can leverage external data without leaving their primary editor

Project-scoped MCP configs (`.vscode/mcp.json`) let you commit server configurations to your repository, ensuring the entire team shares the same tool access

Copilot's Agent mode integrates MCP tools seamlessly with file editing, terminal commands, and workspace search in a single agentic loop

GitHub's enterprise compliance and audit features extend to MCP tool usage, providing visibility into how AI interacts with external services

Braintrust + VS Code Copilot Use Cases

Practical scenarios where VS Code Copilot combined with the Braintrust MCP Server delivers measurable value.

Live API integration: Copilot can query an MCP server, inspect the response schema, and generate typed API client code in the same step

DevSecOps workflows: security teams can give developers access to domain intelligence tools directly in their editor for real-time vulnerability assessment during code review

Data pipeline development: Copilot fetches sample data via MCP and generates transformation scripts, validators, and test fixtures from actual API responses

Documentation generation: Copilot queries available tools and auto-generates README sections, API reference docs, and usage examples

Braintrust MCP Tools for VS Code Copilot (10)

These 10 tools become available when you connect Braintrust to VS Code Copilot via MCP:

create_experiment

Establish a new historical experiment trace to record LLM pipeline tests

create_project

Create a new project environment for tracking AI evaluations and datasets

get_dataset

Retrieve a specific dataset containing exact schemas bounding LLM outputs

get_prompt

Retrieve exact variable contexts and literal text templates for a prompt

insert_dataset_row

Append new test cases into a dataset matrix targeting specific evaluations

list_datasets

List isolated Ground Truth text banks used for automated evaluation scoring

list_env_vars

Probe the Braintrust AI Gateway configurations managing model API keys securely

list_experiments

Retrieve all evaluation experiments mapping model test scores and metrics

list_projects

Retrieve the list of all AI evaluation projects in Braintrust

list_prompts

Retrieve explicitly version-controlled system prompts isolated in Braintrust

Example Prompts for Braintrust in VS Code Copilot

Ready-to-use prompts you can give your VS Code Copilot agent to start working with Braintrust immediately.

"List all active test datasets configured under Braintrust."

"Look up prompt template using specific ID XYZ."

"Analyze recent experiments across multiple models testing behavior."

Troubleshooting Braintrust MCP Server with VS Code Copilot

Common issues when connecting Braintrust to VS Code Copilot through the Vinkius, and how to resolve them.

MCP tools not available

Ensure you are in Agent mode in Copilot Chat. MCP tools only appear in Agent mode.

Braintrust + VS Code Copilot FAQ

Common questions about integrating Braintrust MCP Server with VS Code Copilot.

Which VS Code version supports MCP?

MCP support requires VS Code 1.99 or later with the GitHub Copilot extension. Ensure both are updated to the latest version. Older versions of Copilot may not expose the Agent mode toggle.

How do I switch to Agent mode?

Open the Copilot Chat panel and look for two mode options: "Ask" and "Agent". Click "Agent" to enable autonomous tool calling. In Ask mode, Copilot provides conversational answers but cannot invoke MCP tools.

Can I restrict which MCP tools Copilot can access?

Yes. VS Code shows a tool consent dialog before any MCP tool is invoked for the first time. You can also configure tool access policies at the organization level through GitHub Copilot settings.

Does MCP work in VS Code Remote or Codespaces?

Yes. MCP servers configured via .vscode/mcp.json work in Remote SSH, WSL, and GitHub Codespaces environments. The MCP connection is established from the remote host, so ensure the server URL is accessible from that environment.