mcpField
B
ballpark

ballpark

Local
github.com/CloudCorpRecordsBusiness & Marketing0Updated 1w ago

About ballpark

ballpark is a local MCP bridge and CLI for Dribba’s published studio-data service. It forwards tool calls to the remote endpoint, covering services, case studies, estimates, company facts, and contact requests.

Key features

  1. 01Lists studio services and case studies
  2. 02Estimates project budgets
  3. 03Returns structured company facts
  4. 04Forwards contact requests
  5. 05Passes remote results through unchanged

Use cases

  1. 01Reviewing Dribba’s services and work
  2. 02Estimating an app or product budget
  3. 03Looking up company details
  4. 04Sending a confirmed project inquiry

MCP support

It provides a local stdio MCP server that serves a captured tool list and forwards tool calls to Dribba’s remote MCP endpoint over streamable HTTP.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
"mcpServers": {
"cloudcorprecords-ballpark": {
"command": "npx",
"args": [
"-y",
"@reneturcios/ballpark"
]
}
}
}
Run commandnpx -y @reneturcios/ballpark

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
npm@reneturcios/ballpark0.1.0
ocighcr.io/cloudcorprecords/ballpark:0.1.0

Tools

5

Saved tool list · updated 1w ago

ParameterTypeRequiredDescription
limitintegerNoMaximum number of services to return. Omit for all of them.

Details

From the project README

ballpark

A forwarding client for the published Dribba MCP remote. Dribba is a product studio; its remote MCP server exposes public studio data — its services, its case studies, a ballpark budget calculator, structured company facts, and a contact-request tool. There is no installable package for that remote, so ballpark gives it a local stdio MCP server and a small CLI: tools/list is answered locally from the verbatim live capture of the remote, and every tools/call is forwarded to the remote over streamable HTTP.

The upstream result passes through verbatim, including structuredContent and any isError flag. Failures on the way there or back are reported as an error result whose text starts with ballpark error: .

Upstream and the override

  • Upstream: https://dribba.com/mcp (pinned).
  • Override: set DRIBBA_MCP_URL to point the client at a different endpoint. This override is a reconstruction: no override is published for the remote-only original.
  • The client handshake is initialize (protocol version 2025-03-26, client ballpark / 0.1.0), then notifications/initialized, then tools/call. When the upstream issues an Mcp-Session-Id, it is honoured on every later request. Responses may be plain JSON or an SSE stream.
  • Argument validation is local: an unknown tool or malformed arguments fail before anything is forwarded. submit_contact_request is forwarded like the other tools; its captured description — which asks for explicit user confirmation of the exact name, email and message before calling — is served verbatim, and this rebuild adds no gating of its own.

Install and run

npx -y @reneturcios/ballpark mcp          # stdio MCP server
npx -y @reneturcios/ballpark --version    # 0.1.0
npx -y @reneturcios/ballpark list_services '{"limit":3}'
npx -y @reneturcios/ballpark estimate_project '{"platforms":["ios","web"],"complexity":"standard"}'
npx -y @reneturcios/ballpark list_projects '{"industry":"fintech"}'
npx -y @reneturcios/ballpark company_facts '{"fields":["offices","distinctions"]}'

Each tool subcommand takes one JSON argument string and prints the forwarded result. Exit codes: 0 on success, 1 on an error result, 2 on a usage or validation error.

MCP client config

{
  "mcpServers": {
    "ballpark": {
      "command": "npx",
      "args": ["-y", "@reneturcios/ballpark", "mcp"]
    }
  }
}

Docker

docker run -i ghcr.io/cloudcorprecords/ballpark                    # stdio MCP server
docker run --rm ghcr.io/cloudcorprecords/ballpark --version        # CLI

Tools

ToolWhat it does
list_servicesList Dribba's services with a one-line summary and URL.
estimate_projectBallpark budget and range for an app/product, mirroring dribba.com's public calculator. Minimum project budget is €30,000. Requires platforms.
list_projectsList Dribba's case studies (slug, title, industry).
company_factsStructured facts about Dribba: positioning, offices, team, distinctions, clients.
submit_contact_requestSubmit a real project inquiry to Dribba on behalf of the user. This sends an email and creates a lead — ONLY call it after the user has explicitly confirmed the exact name, email and message to be sent. Dribba replies within 1h. Rate-limited per client.

Development

npm install
npm run build   # tsc -> dist/
npm test        # tsx --test test/*.test.ts (fake fetch only, no network)
npm run dev     # run the CLI from source

License

MIT — Copyright (c) 2026 Rene Turcios.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.