Brain Researcher

Brain Researcher

AI-assisted research infrastructure for neuroimaging

Turn neuroimaging questions into evidence-linked plans, runnable workflows, and reviewable scientific claims.

datasets
1,600+
tool specs
2,000+
MCP tools
87
workflows
60+

Tested across Claude, Codex, Gemini, GLM, DeepSeek, Kimi, and Qwen.

1

Install Claude Code

npm install -g @anthropic-ai/claude-code

npm install -g @anthropic-ai/claude-code
Get Claude Code
2

Get your token

Sign in to mint your personal MCP token (a one-time secret).

Log inCreate account
3

Configure MCP

Export your token, then drop this config into .mcp.json.

# Brain Researcher MCP
export BR_MCP_TOKEN="brk_<kid>.<secret>"
export BR_MCP_AUTH_HEADER="Bearer ${BR_MCP_TOKEN}"
{
  "mcpServers": {
    "brain-researcher": {
      "type": "http",
      "url": "https://brain-researcher.com/mcp",
      "headers": {
        "Authorization": "Bearer ${BR_MCP_TOKEN}",
        "Accept": "application/json, text/event-stream"
      }
    }
  }
}
Try it
Claude alone vs. with Brain Researcher

Now, just ask.

Same question, same agent. Only one of them can find the dataset and hand you a runnable recipe.

your coding agent
›

Claudeno tools
Claude + Brain ResearcherMCP
Streaming demo…
Open Studio
Paper & code
Open & reproducible

Read the paper, run the code

Paper

Brain Researcher: AI-assisted research infrastructure workspace for neuroimaging analyses

Brain Researcher contributors

Preprint pending · arXiv soon

AI agents can write code, run analyses, and propose hypotheses, but a generated output is not a scientific claim by default. Brain Researcher turns researcher judgment into executable commitments: allowed alternatives, validation rules, provenance, and claim boundaries. Every result stays tied to the evidence behind it and the limits beyond which it shouldn't be read. Tested on neuroimaging across seven foundation models, it raised tool-selection accuracy from a 51% to 63% baseline to about 93%, and returned bounded claim records (accepted, qualified, or rejected) instead of unqualified findings.

Code

Brain Researcher

Open-source release coming soon

The full platform: Next.js web UI, FastAPI orchestrator + agent, 87 public MCP tools, 2,000+ registry tool specs, and the Neo4j knowledge graph. We're open-sourcing it so you can wire up your MCP token and reproduce every workflow end to end.

PythonTypeScriptNext.jsFastAPINeo4jMCP
Open in Studio
Demos
See it in action

Explore worked demos

Curated use cases with real evidence bundles, reports, and handoff context. Open one to walk the full research episode.

Browse all demos
Back to top