pyrestoolbox-mcp
MCP serverOfficial (tool developers)
The pyResToolbox author's own MCP server: a few meta-tools that describe and call any public library function. By Mark W. Burgoyne.
98 starsUpdated 5 days agoGitHub data from 5 Nov 2026
At a glance
- Type
- MCP server
- Works with
- pyResToolbox
- How it connects
- Drives a running copy of the tool
- Access
- Read-only
- Tool versions
- pyrestoolbox 3.8.2 or later (server version 0.1.0)
- Operating systems
- Not stated in its documentation
- Needs a pyResToolbox licence
- No
- Transport
- Local (stdio), Remote (HTTP)
- AI clients documented
- Claude Code, Claude Desktop
- Licence
- GPL-3.0-or-later
- Language
- Python
What it can do
- Lists the library's modules and functions, returns a function's signature and docstring, and calls any public function with keyword arguments.
- Adds one-shot tools for correlation recommendations, CO2-brine and multicomponent brine properties, and decline-curve fit and forecast.
- Exposes each module's documentation as an MCP resource, and returns the valid options when an agent passes a wrong correlation name.
- Can run as a stateless remote server over streamable HTTP, with a Dockerfile included.
Limitations
- The author recommends installing the pyrestoolbox Python package directly when your agent can run Python code.
- Not yet published on PyPI as of this check; install from the mcp folder of the repository.
- The oil module accepts scalars only, and table-writing functions return text rather than files.
Install
pip install pyrestoolbox-mcp # once published; or: pip install ./mcpclaude mcp add pyrestoolbox -- pyrestoolbox-mcp{
"mcpServers": {
"pyrestoolbox": {
"command": "pyrestoolbox-mcp"
}
}
}pyrestoolbox-mcp --transport streamable-http --host 0.0.0.0 --port 8000MCP servers and skills run code on your computer. Review the source before installing.
AI-generated models, drawings and results must be checked by a qualified engineer before use.