Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/nvidia/nvalchemi-toolkit/nvalchemi-data-structuresnpx skills add NVIDIA/nvalchemi-toolkit --skill nvalchemi-data-structuresgit clone --depth 1 https://github.com/NVIDIA/nvalchemi-toolkitWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/nvidia/nvalchemi-toolkit/nvalchemi-data-structures)<a href="https://agentmods.dev/skills/nvidia/nvalchemi-toolkit/nvalchemi-data-structures"><img src="https://agentmods.dev/badge/skills/nvidia/nvalchemi-toolkit/nvalchemi-data-structures.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00082 | $0.02837 |
| Opus 5 | $0.00041 | $0.01418 |
| Sonnet 5 | $0.00016 | $0.00567 |
| Haiku 4.5 | $0.00008 | $0.00284 |
Grade A, and why
nvalchemi-data-structures scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 336 lines — stays where its author put it; the contents beside it link to each section on GitHub.
nvalchemi Data Structures
Overview
nvalchemi represents atomic systems as graphs using two core classes:
AtomicData— a single atomic system (molecule, crystal, etc.)Batch— an efficient container of multipleAtomicDataobjects stored as concatenated tensors
Both are Pydantic BaseModel subclasses with DataMixin for device/dtype operations.
from nvalchemi.data import AtomicData, Batch
AtomicData
Construction
Required fields: positions [n_nodes, 3] and atomic_numbers [n_nodes].
import torch
# Minimal
data = AtomicData(
positions=torch.randn(4, 3),
atomic_numbers=torch.tensor([1, 6, 6, 1], dtype=torch.long),
)
# With edges (bonds or neighbor list)
data = AtomicData(
positions=torch.randn(4, 3),
atomic_numbers=torch.tensor([1, 6, 6, 1], dtype=torch.long),
neighbor_list=torch.tensor([[0, 1], [1, 0], [1, 2], [2, 1]], dtype=torch.long),
)
# With system-level fields (energy, cell, pbc)
data = AtomicData(
positions=torch.randn(4, 3),
atomic_numbers=torch.tensor([1, 6, 6, 1], dtype=torch.long),
energy=torch.tensor([[0.5]]),
cell=torch.eye(3).unsqueeze(0), # [1, 3, 3]
pbc=torch.tensor([[True, True, False]]), # [1, 3]
)
From ASE Atoms:
data = AtomicData.from_atoms(
atoms, # ase.Atoms object
energy_key="energy", # key in atoms.info / atoms.calc
forces_key="forces",
device="cpu",
dtype=torch.float32,
)
Field reference
Fields are organized by level. All are optional except positions and atomic_numbers.
| Level | Field | Shape | Notes |
|---|---|---|---|
| Node | atomic_numbers |
[V] |
Required, int64 |
| Node | positions |
[V, 3] |
Required, float |
| Node | atomic_masses |
[V] |
Auto-populated from periodic table |
| Node | atom_categories |
[V] |
Defaults to zeros |
| Node | forces |
[V, 3] |
eV/Angstrom |
| Node | velocities |
[V, 3] |
Auto-initialized to zeros |
| Node | momenta |
[V, 3] |
|
| Node | charges |
[V, 1] |
|
| Node | node_embeddings |
[V, H] |
|
| Node | kinetic_energies |
[V, 1] |
|
| Edge | neighbor_list |
[E, 2] |
COO format, int64 |
| Edge | shifts |
[E, 3] |
Cartesian displacements (neighbor_list_shifts @ cell) |
| Edge | neighbor_list_shifts |
[E, 3] |
Integer lattice image indices |
| Edge | edge_embeddings |
[E, H] |
|
| Dense | neighbor_matrix |
[V, K] |
Dense neighbor matrix (int64) |
| Dense | neighbor_matrix_shifts |
[V, K, 3] |
Periodic shifts for dense neighbors |
| Dense | num_neighbors |
[V] |
Valid neighbor count per atom |
| System | cell |
[1, 3, 3] |
Lattice vectors |
| System | pbc |
[1, 3] |
Periodic boundary conditions (bool) |
| System | energy |
[1] |
eV |
| System | stress |
[1, 3, 3] |
eV/Angstrom^3 |
| System | virial |
[1, 3, 3] |
|
| System | dipole |
[1, 3] |
|
| System | charge |
[1] |
|
| System | graph_embeddings |
[1, H] |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 4d ago First seen · 336 lines · 82 tokens per session scan A bc950235e050
nvalchemi-data-structures is a skill published in the GitHub repository NVIDIA/nvalchemi-toolkit (150 stars, last pushed 9d ago), licensed Apache-2.0. It adds 82 tokens to every session and 2,837 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
geomaster
Comprehensive geospatial science skill covering remote sensing, GIS, spatial analysis, machine learning for earth observation, and 30+ scientific domains. Supports satellite imagery processing (Sentinel, Landsat, MODIS, SAR, hyperspectral), vector and raster data operations, spatial statistics, point cloud processing…
evaluating-with-leakage-gates
Evaluate an OpenMed de-identification or clinical NER model against the leakage-first release gates G1a through G8, which gate releases on residual PHI leakage rather than on F1. Use when the user wants to run the OpenMed eval harness on a synthetic golden set, decide whether a de-id model is RELEASABLE or…
mapping-loinc
Maps laboratory and clinical observation names extracted by OpenMed to LOINC codes using the public Regenstrief LOINC and FHIR terminology APIs. Use when the user wants to code lab tests, vital signs, or observations to LOINC, resolve a test name plus specimen and method to the correct LOINC part-model code, attach…
pysr
Use when fitting equations to data with PySR or SymbolicRegression.jl, when a user wants an interpretable formula, symbolic model, scaling law, or empirical relation discovered from numeric data, or when debugging a PySR search that is slow, stuck, or giving poor equations.
evo2
Score, embed, and generate DNA sequences with Evo 2, a long-context genomic foundation model. Use this skill when: (1) Computing per-nucleotide or per-sequence likelihoods for variant effect scoring, (2) Embedding genomic windows for downstream classification, (3) Generating DNA conditioned on a prefix, (4) Scoring…
experimental-design
Best practices for designing reproducible ML experiments. Use when planning ablations, baselines, or controlled experiments.