PymatgenSAFE

Kosmos: An AI Scientist for Autonomous Discovery - An implementation and adaptation to be driven by Claude Code or API - Based on the Kosmos AI Paper - https://arxiv.org/abs/2511.02824

The instruction file, verbatim from the audited commit — this is the text the model reads, and the surface the audit's instruction layer examines. Quoted here so you can judge it without cloning anything.

---
name: pymatgen
description: "Materials science toolkit. Crystal structures (CIF, POSCAR), phase diagrams, band structure, DOS, Materials Project integration, format conversion, for computational materials science."
---

# Pymatgen - Python Materials Genomics

## Overview

Pymatgen is a comprehensive Python library for materials analysis that powers the Materials Project. Create, analyze, and manipulate crystal structures and molecules, compute phase diagrams and thermodynamic properties, analyze electronic structure (band structures, DOS), generate surfaces and interfaces, and access Materials Project's database of computed materials. Supports 100+ file formats from various computational codes.

## When to Use This Skill

This skill should be used when:
- Working with crystal structures or molecular systems in materials science
- Converting between structure file formats (CIF, POSCAR, XYZ, etc.)
- Analyzing symmetry, space groups, or coordination environments
- Computing phase diagrams or assessing thermodynamic stability
- Analyzing electronic structure data (band gaps, DOS, band structures)
- Generating surfaces, slabs, or studying interfaces
- Accessing the Materials Project database programmatically
- Setting up high-throughput computational workflows
- Analyzing diffusion, magnetism, or mechanical properties
- Working with VASP, Gaussian, Quantum ESPRESSO, or other computational codes

## Quick Start Guide

### Installation

```bash
# Core pymatgen
uv pip install pymatgen

# With Materials Project API access
uv pip install pymatgen mp-api

# Optional dependencies for extended functionality
uv pip install pymatgen[analysis]  # Additional analysis tools
uv pip install pymatgen[vis]       # Visualization tools
```

### Basic Structure Operations

```python
from pymatgen.core import Structure, Lattice

# Read structure from file (automatic format detection)
struct = Structure.from_file("POSCAR")

# Create structure from scratch
lattice = Lattice.cubic(3.84)
struct = Structure(lattice, ["Si", "Si"], [[0,0,0], [0.25,0.25,0.25]])

# Write to different format
struct.to(filename="structure.cif")

# Basic properties
print(f"Formula: {struct.composition.reduced_formula}")
print(f"Space group: {struct.get_space_group_info()}")
print(f"Density: {struct.density:.2f} g/cm3")
```

### Materials Project Integration

```bash
# Set up API key
export MP_API_KEY="your_api_key_here"
```

```python
from mp_api.client import MPRester

with MPRester() as mpr:
    # Get structure by material ID
    struct = mpr.get_structure_by_material_id("mp-149")

    # Search for materials
    materials = mpr.materials.summary.search(
        formula="Fe2O3",
        energy_above_hull=(0, 0.05)
    )
```

## Core Capabilities

### 1. Structure Creation and Manipulation

Create structures using various methods and perform transformations.

**From files:**
```python
# Automatic format detection
struct = Structure.from_file("structure.cif")
struct = Structure.from_file("POSCAR")
mol = Molecule.from_file("molecule.xyz")
```

**From scratch:**
```python
from pymatgen.core import Structure, Lattice

# Using lattice parameters
lattice = Lattice.from_parameters(a=3.84, b=3.84, c=3.84,
                                  alpha=120, beta=90, gamma=60)
coords = [[0, 0, 0], [0.75, 0.5, 0.75]]
struct = Structure(lattice, ["Si", "Si"], coords)

# From space group
struct = Structure.from_spacegroup(
    "Fm-3m",
    Lattice.cubic(3.5),
    ["Si"],
    [[0, 0, 0]]
)
```

**Transformations:**
```python
from pymatgen.transformations.standard_transformations import (
    SupercellTransformation,
    SubstitutionTransformation,
    PrimitiveCellTransformation
)

# Create supercell
trans = SupercellTransformation([[2,0,0],[0,2,0],[0,0,2]])
supercell = trans.apply_transformation(struct)

# Substitute elements
trans = SubstitutionTransformation({"Fe": "Mn"})
new_struct = trans.apply_transformation(struct)

# Get primitive cell
trans = PrimitiveCellTransformation()
primitive = trans.apply_transformation(struct)
```

**Reference:** See `references/core_classes.md` for comprehensive documentation of Structure, Lattice, Molecule, and related classes.

### 2. File Format Conversion

Convert between 100+ file formats with automatic format detection.

**Using convenience methods:**
```python
# Read any format
struct = Structure.from_file("input_file")

# Write to any format
struct.to(filename="output.cif")
struct.to(filename="POSCAR")
struct.to(filename="output.xyz")
```

**Using the conversion script:**
```bash
# Single file conversion
python scripts/structure_converter.py POSCAR structure.cif

# Batch conversion
python scripts/structure_converter.py *.cif --output-dir ./poscar_files --format poscar
```

**Reference:** See `references/io_formats.md` for detailed documentation of all supported formats and code integrations.

### 3. Structure Analysis and Symmetry

Analyze structures for symmetry, coordination, and other properties.

**Symmetry analysis:**
```python
from pymatgen.symmetry.analyzer import SpacegroupAnalyzer

sga = SpacegroupAnalyzer(struct)

# Get space group information
print(f"Space group: {sga.get_space_group_symbol()}")
print(f"Number: {sga.get_space_group_number()}")
print(f"Crystal system: {sga.get_crystal_system()}")

# Get conventional/primitive cells
conventional = sga.get_conventional_standard_structure()
primitive = sga.get_primitive_standard_structure()
```

**Coordination environment:**
```python
from pymatgen.analysis.local_env import CrystalNN

cnn = CrystalNN()
neighbors = cnn.get_nn_info(struct, n=0)  # Neighbors of site 0

print(f"Coordination number: {len(neighbors)}")
for neighbor in neighbors:
    site = struct[neighbor['site_index']]
    print(f"  {site.species_string} at {neighbor['weight']:.3f} Å")
```

**Using the analysis script:**
```bash
# Comprehensive analysis
python scripts/structure_analyzer.py POSCAR --symmetry --neighbors

# Export results
python scripts/structure_analyzer.py structure.cif --symmetry

SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.

What the source does

Filesystem

none-observed

Network

none-observed

Shell

none-observed

Dependencies

pinned

Secrets in source

none-found

Findings (0)

No findings outside the package's declared scope.

Gates applied: no_behavioural_pass.