Skip to main content
Glama
CoolProp MCP Server

CoolProp MCP Server

by pipeworx-io

@pipeworx/coolprop

Thermophysical properties of a pure fluid, computed from two independent state variables (PropsSI-style: give e.g. temperature + pressure, get density, enthalpy, entropy, viscosity, thermal conductivity, specific heat, vapor quality, phase). Covers ~124 fluids — water, common refrigerants, hydrocarbons, cryogens, siloxanes and more — computed in-process by CoolProp rather than looked up from a table, so a query is valid anywhere inside each fluid's full equation-of-state range, not just at tabulated points.

Part of Pipeworx — an MCP gateway connecting AI agents to 1863+ live data sources. This is an independent, unofficial integration — not affiliated with, endorsed by, or published by the upstream provider.

Tools

  • coolprop_fluid_properties(fluid, input1_name, input1_value, input2_name, input2_value, outputs) — two independent state variables (T, P, D, H, S, U, or Q, SI units) in, up to 12 requested properties out: T, P, D (density, kg/m3), H (enthalpy, J/kg), S (entropy, J/kg/K), U (internal energy, J/kg), Q (vapor quality), CPMASS/CVMASS (specific heats, J/kg/K), VISCOSITY (Pa·s), CONDUCTIVITY (W/m/K), PRANDTL, M (molar mass, kg/mol), Phase (liquid/gas/supercritical/twophase/...).

  • coolprop_list_fluids() — the exact ~124 canonical fluid names this build recognizes. No aliases: use "CarbonDioxide", not "CO2" — this tool is how you find the exact spelling rather than guessing one.

Related MCP server: MoziChem-Hub MCP Server

Auth

Keyless.

Data sources

  • CoolProp (Ian H. Bell and the CoolProp developers) — an open-source thermophysical property library (MIT licence). Every call computes the property in-process from CoolProp's own equations of state; nothing is looked up from a pre-built table of values, and nothing is fetched from a live upstream.

Notes

  • No aliases. This build's fluid list carries only canonical names (e.g. "CarbonDioxide", "Nitrogen", "R134a") — common short forms like "CO2" are not recognized and return a clear error rather than a wrong answer. Call coolprop_list_fluids to get the exact name.

  • input1_name and input2_name must be two different, independent state variables. Asking for T twice (or any other repeated/dependent pair) returns a named error rather than a nonsense number.

  • A state CoolProp cannot resolve returns a named error, not a silent wrong value. CoolProp's own low-level convention for an unresolvable state is the IEEE Infinity sentinel (not an exception) — this pack detects that and reports it as an error naming the fluid, inputs and valid input codes, rather than ever returning Infinity as though it were a real density or enthalpy.

  • A small number of (fluid, state) combinations fail to resolve even when the state is physically unremarkable (observed live for pure CarbonDioxide across a range of atmospheric-pressure temperatures) — this reads as the same named error as an out-of-range state, since there is no way to distinguish a genuine CoolProp solver gap from an invalid input at the property-call level. If a fluid you expect to work returns this error, try a different input pair (e.g. density + temperature instead of pressure + temperature) before concluding the state is invalid.

  • Mass-basis properties only (no molar-basis inputs/outputs in this version).

Quick Start

Add to your MCP client (Claude Desktop, Cursor, Windsurf, etc.):

{
  "mcpServers": {
    "coolprop": {
      "url": "https://gateway.pipeworx.io/coolprop/mcp"
    }
  }
}

What this endpoint actually serves

tools/list at https://gateway.pipeworx.io/coolprop/mcp returns the tools in the table above plus the shared Pipeworx meta-tools — ask_pipeworx, discover_tools, search_within, remember/recall and the rest of the gateway-wide set. So the tool count you see is larger than this table: a single-pack endpoint currently lists roughly 30 shared tools alongside the pack's own. The connection's initialize response states its exact scope, and is the authoritative answer for a given day.

This is deliberate, not multiplexing by accident. The meta-tools are what let a scoped connection answer a question this pack does not cover — via ask_pipeworx, which routes across the whole catalog — without you adding a second MCP server. There is currently no way to mount a pack endpoint without them; if the extra schemas cost you more context than the routing is worth, connect to the full gateway once rather than to several pack endpoints.

Or connect to the full Pipeworx gateway to get every pack's tools listed directly, instead of just this one's:

{
  "mcpServers": {
    "pipeworx": {
      "url": "https://gateway.pipeworx.io/mcp"
    }
  }
}

Both URLs reach the same gateway and the same 1863+ data sources. The only difference is which pack's tools are listed directly; ask_pipeworx reaches all of them from either one.

No MCP client? Call it over HTTP

curl -X POST https://gateway.pipeworx.io/v1/tools/coolprop_fluid_properties \
  -H 'Content-Type: application/json' \
  -d '{"fluid":"Water","input1_name":"T","input1_value":298.15,"input2_name":"P","input2_value":101325,"outputs":["D","H","S","VISCOSITY","CONDUCTIVITY","Phase"]}'

No account needed for the first calls. Inspect any tool: GET https://gateway.pipeworx.io/v1/tools/coolprop_fluid_properties. Find one: POST https://gateway.pipeworx.io/v1/tools/search_packs with {"query":"..."}.

Run it from npm

@pipeworx/mcp-coolprop is a local stdio bridge to the Pipeworx gateway: your MCP client starts it with npx, and it forwards every call to the same https://gateway.pipeworx.io/coolprop/mcp endpoint shown above. It needs network access; there is no offline mode.

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

Without a key it calls anonymously, capped at 50 calls a day. A free account raises that to 200 a day: get a key at https://pipeworx.io/signup and add "env": { "PIPEWORX_API_KEY": "<your key>" } to the server entry. The key is sent as a Bearer token, the same way a hosted connection sends it, so calls are metered and billed exactly as they are there.

It lists only this pack's tools by default. Set PIPEWORX_META_TOOLS=1 to also list ask_pipeworx and the other gateway-wide tools.

Behind an HTTPS proxy: Node's built-in fetch does not read HTTPS_PROXY by default. Newer Node releases honour it when NODE_USE_ENV_PROXY=1 is set; check your Node version's documentation.

Using with ask_pipeworx

Instead of calling tools directly, you can ask questions in plain English — this works on the pack endpoint above as well as on the full gateway:

ask_pipeworx({ question: "your question about Coolprop data" })

The gateway picks the right tool and fills the arguments automatically.

More

License

MIT

Related MCP Connectors

Related MCP Servers

  • A
    license
    Not graded
    quality
    B
    maintenance
    A physics engine for liquid-cooled GPU systems, exposed as an AI-callable MCP server. Enables thermal analysis, coolant comparison, flow optimization, and rack-level sizing via natural language queries.
    1
    MIT
  • F
    license
    Not graded
    quality
    D
    maintenance
    MCP server exposing thermodynamic and transport properties of refrigerants and ~130 other fluids via CoolProp, enabling queries of state points, saturation, fluid info, and humid air properties through natural language.
    -