How to add ENCODE Toolkit to Cursor

Search ENCODE, cross-reference 14 genomics databases, run analysis pipelines, and generate publication-ready methods from natural language. Paste the config into ~/.cursor/mcp.json and restart Cursor.

Last updated June 14, 2026 · 35★ · stdio · no auth

Cursor config for ENCODE Toolkit

claude mcp add encode -- uvx encode-toolkit

{
  "mcpServers": {
    "encode-toolkit": {
      "command": "uvx",
      "args": [
        "encode-toolkit"
      ]
    }
  }
}

Requires `uv` (the Python package runner). Install it from https://docs.astral.sh/uv/ if `uvx` is not found.

Setup steps

1Open Cursor → Settings → MCP → Add new MCP server (or edit ~/.cursor/mcp.json directly).
2Paste the ENCODE Toolkit config below into the "mcpServers" object.
3Fill in placeholder secrets, then save.
4Cursor reloads MCP servers automatically — check Settings → MCP for a green status dot.
5Ask Cursor to use one of ENCODE Toolkit's tools to confirm it's connected.

Before you start

uvx (from uv) or Node.js/npx, or pip — to run the encode-toolkit package
An MCP-compatible client (Claude Code, Claude Desktop, Cursor, Windsurf, or VS Code)
Optional: ENCODE account access keys for restricted/unreleased data

What ENCODE Toolkit can do in Cursor

encode_search_experiments

Search ENCODE experiments with 20+ filters (assay title, organism, organ, biosample type, target, biosample term name, limit).

encode_get_experiment

Get full details for a single experiment by accession, including all files, quality metrics, and audit info.

encode_download_files

Download specific files by accession to a local directory, with organization options and optional MD5 verification.

encode_batch_download

Search and download in one step; runs in preview (dry_run) mode by default until disabled.

encode_track_experiment

Track an experiment locally along with its publications, methods, and pipeline info.

encode_list_files

List files for a specific experiment with format, output type, assembly, and preferred-default filters.

encode_search_files

Search files across all experiments combining experiment-level and file-level filters.

encode_get_metadata

List valid filter values for any parameter (assays, organisms, organs, biosample types, file formats, output types, assemblies).

Security

100% local execution with no telemetry or analytics. Optional ENCODE credentials for restricted/unreleased data are encrypted at rest via the OS keyring (macOS Keychain, Linux Secret Service, Windows Credential Locker) with a Fernet fallback. Certificate verification is enforced (no verify=False), downloads are MD5-verified by default, and requests are rate limited to ENCODE's 10 req/sec policy. No data leaves the machine except queries to public APIs over HTTPS.

ENCODE Toolkit + Cursor FAQ

Where is the Cursor config file?

Cursor reads MCP servers from ~/.cursor/mcp.json. Paste the ENCODE Toolkit config there under the "mcpServers" key and restart the client.

Is ENCODE Toolkit safe to use with Cursor?

Do I need ENCODE credentials to use it?

No. Most ENCODE data is public and requires no authentication. Credentials are only needed for restricted or unreleased data; when provided they are encrypted at rest in your OS keyring.

Can I install just the MCP server without the skills?

Yes. Run `claude mcp add encode -- uvx encode-toolkit` (or use npx/pip) to install only the 20 MCP tools. The 47 workflow skills and 7 pipelines come with the Claude Code plugin install.

Is it affiliated with the ENCODE Consortium?

No. It is an independent open-source toolkit by Dr. Alex M. Mawla that accesses the public ENCODE Project API; it is not an official ENCODE Consortium product.

View repo Full ENCODE Toolkit page