openmc-validator-mcp
An MCP server that provides pre-flight validation and post-processing tools for OpenMC Monte Carlo transport simulations, catching common authoring mistakes before jobs hit the HPC queue.
README
openmc-validator-mcp
An MCP (Model Context Protocol) server that provides pre-flight validation and post-processing tools for OpenMC Monte Carlo transport simulations. Catch common authoring mistakes before the job hits the HPC queue — eliminate the costly "submit → queue → fail → fix → resubmit" loop.
Features
The server exposes 14 tools across four categories:
Validators (5) — Catch mistakes at authoring time
| Tool | What it checks |
|---|---|
validate_materials |
Nuclide name typos, negative density, ao/wo fraction mixing, L2 round-trip |
validate_geometry |
Unclosed surface references, unknown universes/cells, DAGMC path validity |
validate_settings |
Batch/population sanity, source definition completeness, D-T energy check |
validate_tallies |
Energy bin ordering, mesh axis validity, score whitelist |
validate_model |
End-to-end model validation; optional openmc --geometry-debug subprocess |
Cross-Checks (3) — Spot interface mismatches
| Tool | What it checks |
|---|---|
check_cross_section_library |
Nuclide + temperature availability in cross_sections.xml |
check_energy_filter_coverage |
Energy filter upper edge vs. source max energy (e.g. 14.1 MeV) |
check_tally_filter_consistency |
Tally cell filters vs. geometry cells, mesh extent boundaries |
Converters (2) — Bridge OpenMC to downstream codes
| Tool | What it does |
|---|---|
convert_spectrum_openmc_to_fispact |
OpenMC energy tally → FISPACT-II ARBI spectrum file |
extract_tally_summary |
statepoint.h5 → structured JSON (mean / std / rel_err) |
Template & Material Catalog (4) — Browse pre-built OpenMC templates
| Tool | What it does |
|---|---|
list_templates |
List all available OpenMC model templates by category |
get_template |
Retrieve a specific template with all files (geometry, materials, settings) |
list_materials |
List all pre-defined material definitions across templates |
get_material |
Look up a specific material by name, get nuclide composition |
Requires OPENMC_TEMPLATES_DIR environment variable pointing to an
openmc-templates repository.
Error Code System
40+ structured error codes across 7 domains (MAT-*, SET-*, TAL-*, GEO-*,
XS-*, EFC-*, TFC-*, MOD-*, ENV-*), each with severity, trigger
description, and suggested fix. See docs/error-codes.md.
Installation
Requires Python ≥ 3.10.
# Basic install (L1 validation only, no OpenMC API required)
pip install openmc-validator
# Editable install for development
pip install -e ".[dev]"
For L2 validation (material round-trip through OpenMC API, geometry-debug):
# OpenMC is only available via conda-forge
conda create -n openmc-env -c conda-forge python=3.11 openmc
conda activate openmc-env
pip install -e ".[dev]"
Without OpenMC, L2-dependent checks degrade gracefully with ENV-W001 warnings.
Quick Start
As an MCP Server
Add to your VS Code mcp.json:
{
"servers": {
"openmc-validator": {
"command": "openmc-validator",
"args": []
}
}
}
See examples/mcp-config.json for a complete
example with OPENMC_CROSS_SECTIONS env.
Programmatic Use
from openmc_validator.validators.materials import validate_materials
from openmc_validator.schemas import MaterialSpec, NuclideSpec
spec = MaterialSpec(
name="Water",
density=1.0,
density_units="g/cm3",
nuclides=[
NuclideSpec(name="H1", fraction=2.0, fraction_type="ao"),
NuclideSpec(name="O16", fraction=1.0, fraction_type="ao"),
],
)
report = validate_materials([spec])
print(report.passed, report.issues)
Documentation
| Document | Description |
|---|---|
docs/tool-reference.md |
Full API reference for all 14 tools |
docs/error-codes.md |
Index of all validation error/warning codes |
docs/changelog.md |
Version history |
examples/agent-integration.md |
Using the server with AI coding agents |
Project Layout
src/openmc_validator/
schemas/ # pydantic input/output models
validators/ # 5 validator modules
crosschecks/ # 3 cross-check modules
converters/ # 2 converter modules
utils/ # subprocess helpers, logging
server.py # FastMCP entry point
tests/
unit/ # pure-python unit tests (no openmc binary required)
integration/ # end-to-end MCP protocol tests
fixtures/
valid/ # positive examples
invalid/ # negative examples (expected to fail validation)
Contributing
Contributions are welcome! Please see CONTRIBUTING.md for
development setup, test instructions, and pull request guidelines.
This project follows the Contributor Covenant Code of
Conduct. Security vulnerabilities should be reported as described in
SECURITY.md.
License
MIT — see LICENSE for details.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.