← Registry

AI & Machine Learning

bigcompute.science

Manages machine learning experiments, datasets, and research problems.

1 endpoint23 known toolsFirst detected May 11, 2026Last detected August 23, 2026

ENDPOINT 1

https://mcp.bigcompute.science/mcp

No auth detected

MCP server metadata

Name
bigcompute.science
Version
1.0.0
Capabilities
toolsresources

Known tools 23

list_experiments

List all experiments with status and key results

Inferred read-only
get_experiment

Get full details of a specific experiment

Inferred read-only
get_zaremba_exceptions

Get the 27 Zaremba exceptions for A={1,2,3}

Inferred read-only
list_datasets

List all Hugging Face datasets

Inferred read-only
get_open_problems

Get open problems that need GPU compute

Inferred read-only
get_cuda_kernel

Get CUDA kernel compile and run commands

Inferred read-only
search

Search experiments, findings, and datasets

Inferred read-only
search_arxiv

Search arXiv for math/CS papers

Inferred read-only
search_papers

Search academic papers via Semantic Scholar with citation counts

Inferred read-only
lookup_oeis

Look up integer sequences in OEIS by terms or keywords

Inferred read-only
lookup_lmfdb

Query LMFDB for number fields, elliptic curves, modular forms

Inferred read-only
search_zbmath

Search zbMATH Open — 4.5M+ reviewed math publications with MSC codes

Inferred read-only
search_mathlib

Search Lean 4 Mathlib theorems via Loogle by type signature or name

Inferred read-only
search_findstat

Search FindStat combinatorial statistics database

Inferred read-only
verify_finding

Cross-reference a finding against arXiv, Semantic Scholar, zbMATH, OEIS

Inferred read-only
get_finding

Get full finding details with data and references (no live API calls)

Inferred read-only
list_findings

List all published findings with claims and data

Inferred read-only
suggest_experiment

Get recommended next experiments based on GPU time and interest area

Inferred read-only
get_certification_process

Explains MCP Verification Certification levels and how agents can contribute

Inferred read-only
search_boise_state

Search Boise State University ScholarWorks repository

Inferred read-only
search_fau

Search Florida Atlantic University digital library

Inferred read-only
get_research_agent

Get instructions to run the autonomous research agent — fastest way to contribute

Inferred read-only
list_related_servers

Discover other MCP servers for math and academic research

Inferred read-only

CONNECT WITH APPROVAL

Client installation

Review this server and its permissions before adding it. Secret placeholders must be set locally.

Codex

~/.codex/config.toml

[mcp_servers.bigcompute-science]
url = "https://mcp.bigcompute.science/mcp"
enabled = true
Claude Code

.mcp.json

{
  "mcpServers": {
    "bigcompute-science": {
      "type": "http",
      "url": "https://mcp.bigcompute.science/mcp"
    }
  }
}
Claude Desktop

Settings → Connectors → Add custom connector

Name: bigcompute-science
Remote MCP URL: https://mcp.bigcompute.science/mcp

Add this remote URL as a custom connector in Claude Desktop. Availability depends on the user plan and workspace policy.

Cursor

.cursor/mcp.json

{
  "mcpServers": {
    "bigcompute-science": {
      "url": "https://mcp.bigcompute.science/mcp"
    }
  }
}
Visual Studio Code

.vscode/mcp.json

Add to Visual Studio Code
{
  "servers": {
    "bigcompute-science": {
      "type": "http",
      "url": "https://mcp.bigcompute.science/mcp"
    }
  }
}
Generic MCP

Client-specific MCP configuration

{
  "name": "bigcompute-science",
  "transport": "streamable-http",
  "url": "https://mcp.bigcompute.science/mcp"
}
MCP Inspector

Run the official MCP Inspector locally and enter the indexed Streamable HTTP endpoint.

TRUST AND VERIFICATION EVIDENCE

Trust Data Available

BuiltWith Trust API v2 evidence for bigcompute.science was fetched 2026-08-25T02:28:02.893Z.

Trust status Neutral

bigcompute.science is assessed as Neutral: No suspicious signals found, but no strong positive signal either

Indexed

Evidence is source-attributed and does not guarantee that a third-party server is safe. Risk labels are conservative metadata heuristics.