"Fortran" matching MCP servers:

Enables AI models to write, format, compile, lint, and structure modern Fortran code using best practices and industry design patterns.

Serves a facts-only index of SWAT+ Fortran source as read-only MCP tools, enabling coding assistants to answer exhaustive questions about loops, call graphs, file I/O, and other source facts with file and line references.

Serves AI agents live, non-RAG access to registered HEP codebases through MCP tools for regex search, symbol lookup, and file reading across C++, Python, Fortran, and Mathematica sources.

nativegate is a code generator that produces MCP servers from legacy native codebases: point it at a Fortran deck or C++ header and it generates pybind11/f2py bindings, a Python package, a FastAPI service that exposes the native routines as MCP tools and REST endpoints, tests, Docker/Kubernetes manifests, and a numerical golden-record regression harness. The legacy numerics are never rewritten

Standalone MCP server for code structure analysis using tree-sitter. Directory trees, symbol definitions, and call graphs without reading raw source files. Supports Rust, Python, Go, Java, TypeScript, Fortran, JavaScript, C/C++, and C#. Benchmarked up to 68% fewer tokens vs native tools.