Built by Apeiron Intelligence  ·  Runs on JSI 🐙

AI meets Materials Science

Specialized MCP servers that give LLMs the ability to run quantum simulations, analyze material properties, and access structured knowledge — reducing token usage and dramatically increasing accuracy.

Explore MCP servers About Apeiron Intelligence

🪸 MCP Reef

Purpose-built tool servers for computational materials science, designed to plug into JSI and any MCP-compatible platform.

🐠

MagicDorye

In development

A unified framework for materials simulation that wraps major computational engines into a single, interoperable interface. Built on Pydantic objects and actions architecture — MagicDorye lets AI agents set up, run, and parse atomistic simulations without guessing parameters. Integrated with LSQuant for multiscale workflows.

Quantum ESPRESSO — SCF, band structure, DOS calculations with automatic input generation
Siesta — DFT calculations, band structure, SCF setup for large-scale systems
Structure actions — VASP format conversion, crystal manipulation, atomic structure building
Phonopy integration — phonon dispersion, thermal properties
TB2J + Vampire — magnetic exchange parameters, spin dynamics simulations
LSQuant integration — multiscale workflows from DFT to tight-binding Hamiltonians
Result parsing — extract energies, forces, band gaps from QE and Siesta output
Client-server architecture — Docker-based, distributed calculations via REST API bridge
Visualization — crystal structures, band diagrams, scanning probe simulations
🏛️

Atlantis

Planned

A graph knowledge base for materials science — a structured, queryable repository of material properties, crystal structures, simulation parameters, and literature references. Acts as an extension of JSI's Alexandria memory system, but specialized for the physical sciences.

Graph database of materials, properties, and relationships
Query by composition, structure type, space group, or property range
Validated simulation parameters — cutoffs, k-meshes, pseudopotentials per material
Literature references — link properties to published papers and datasets
Auto-suggest parameters for MagicDorye — "what settings should I use for silicon DFT?"
Community contributions — users can submit validated results that enrich the graph

Example: band gap calculation in chat

👤

User

"Calculate the band gap of GaN in the wurtzite structure using DFT."

🐙

JSI Master

Plans the workflow: build structure → query knowledge base for parameters → run QE → parse results. Delegates to workers with MagicDorye tools.

🏛️

Atlantis MCP

Returns validated pseudopotentials, recommended cutoff (80 Ry), k-mesh (8×8×4), and lattice parameters from literature.

🐠

MagicDorye MCP

Generates input files → runs SCF relaxation → runs band structure calculation → parses output. Returns: band gap = 1.71 eV (GGA-PBE), with convergence data.

🐙

JSI Master

"The calculated band gap of wurtzite GaN is 1.71 eV (PBE), which underestimates the experimental value of 3.4 eV — typical for GGA. Want me to run a hybrid functional (HSE06) for a more accurate result?"