breedsim-mcp server

breedsim-mcp

Breeding-scheme simulation over MCP — returns distributions, never a single stochastic run.

Drives AlphaSimR so an agent can ask what a selection programme would actually gain, with one structural rule: a single simulation run is not a result, and this API will not return one.

Measured on AlphaSimR 2.1.0 — five seeds of an identical three-cycle programme gave mean genetic gain [1.151, 1.841, 1.424, 1.429, 1.473]: sd 0.247 on the very number being reported. Quoting one run to three decimals reports noise with the authority of a measurement. So run_program enforces a replicate floor and returns per-cycle mean, sd and confidence interval. There is no flag that collapses it to a point estimate.

Unofficial. Not affiliated with, endorsed by, or sponsored by the AlphaSimR authors, the

University of Edinburgh, or the R Foundation. See NOTICE.

<!-- mcp-name: io.github.musharna/breedsim-mcp -->

Status

On PyPI — the badge above is the released version, so it cannot go stale the way a number typed here would. Genomic selection included. 5 tools, 57 tests against real AlphaSimR, and 20 mutation checks all confirmed red (docs/MUTATION-CHECKS.md). The simulation is also checked against the breeder's equation R = h²S rather than only against itself (docs/EVAL.md). CI installs R and compiles AlphaSimR, so the suite runs against the real engine on Python 3.11, 3.12 and 3.13 — not against a mock.

Requires mcp 2.x.

Install tax — read this first

Heavier than uv pip install, and the reasons are not negotiable:

• R ≥ 4.3 with a shared library (libR.so)
• AlphaSimR — an Rcpp/RcppArmadillo compile, minutes not seconds
• libtirpc-dev — rpy2 fails to link without it (cannot find -ltirpc)
• rpy2 pinned <3.6 — 3.6 binds R_getVar, which needs R ≥ 4.4

sudo apt-get install -y r-base r-base-dev libtirpc-dev
R -e 'install.packages("AlphaSimR", repos="https://cloud.r-project.org")'
uv add breedsim-mcp

If you build against a conda Python, rpy2 will fail to load libR.so with GLIBCXX_3.4.30 not found — conda ships an older libstdc++ than system libicuuc requires. Use a system or uv-managed interpreter.

Configure your MCP client

The server speaks stdio; the installed console script is breedsim-mcp.

Claude Code

claude mcp add breedsim -- breedsim-mcp

Claude Desktop — add to claudedesktopconfig.json:

{
  "mcpServers": {
    "breedsim": {
      "command": "breedsim-mcp"
    }
  }
}

If the executable is not on your PATH, or AlphaSimR lives in a user library, invoke it through uv and pass the library path:

{
  "mcpServers": {
    "breedsim": {
      "command": "uv",
      "args": ["run", "--directory", "/path/to/breedsim-mcp", "breedsim-mcp"],
      "env": { "R_LIBS_USER": "/home/you/R/library" }
    }
  }
}

Verify with listmethods(), which reports the engine versions and whether this process can currently produce reproducible results**.

Tools

Shortened. The full README is on GitHub.