Netneurotools Guide

What it does

When your agent uses it

• The user works with brain connectivity matrices
• Graph theory on brain networks
• Parcellated brain data (Schaefer
• Desikan-Killiany)

Example prompts

• “/netneurotools-guide”

Requirements

• Python 3

Workflow steps

12 steps, taken from the first numbered list in SKILL.md.

1. 1SURFACE fields are .L and .R, not .lh and .rh. Use surface.L and surface.R to access hemisphere paths.
2. 2Weight-to-distance conversion must be done before calling distance_wei_floyd or search_information. Common transform: D = -np.log(W /…
3. 3Minimum permutation p-value is 1 / (n_perm + 1). With n_perm=1000, the smallest p-value is ~0.001.
4. 4consensus_modularity requires non-negative input. Louvain cannot handle negative weights. Set negatives to zero: A[A < 0] = 0.
5. 5struct_consensus hemiid encoding: 0 = right hemisphere, 1 = left hemisphere.
6. 6pv_plot_surface data format: When hemi='both', vertex_data can be a tuple (left, right) or a single concatenated array. Data length must…
7. 7Data directory: All fetch functions default to ~/nnt-data. Override with data_dir= parameter or set NNT_DATA environment variable.
8. 8match_length_degree_distribution recommended nswap: Use nswap = nnodes * 20 for adequate randomization.
9. 9strength_preserving_rand_sa frac parameter must be between 0 and 1. It controls temperature decrease per annealing stage.
10. 10parcels_to_vertices and vertices_to_parcels support .annot, .gii, and .dlabel.nii parcellation files. For .dlabel.nii, pass a single file…
11. 11Numba acceleration is available for spatial stats, weighted correlation, and some metrics. Install numba for significant speedups on large…
12. 12Headless rendering with PyVista: set os.environ["VTK_DEFAULT_OPENGL_WINDOW"] = "vtkOSOpenGLRenderWindow" before importing pyvista.

What it can do on your machine

Read from SKILL.md and the folder at commit 93f6855. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  Shell commands in SKILL.md call:

  • pip

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  No URLs in SKILL.md. Its commands use pip, which can reach the network depending on how they are called.

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Netneurotools Guide loads about 2.6k tokens when it runs, and up to ~23k if it reads all its reference files. Until then it costs about 214 tokens; SKILL.md has 633 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~214

When it runs · the whole SKILL.md, loaded when a task matches

~2.6k

With references · SKILL.md plus every file in references/, read only if the agent opens them

~23k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from NeuroAIHub/BrainPilot at commit 93f6855, republished under its AGPL-3.0 licence (© NeuroAIHub). 633 words, ~2,553 tokens.

name

netneurotools-guide

description

Domain-validated guidance for network neuroscience analysis using netneurotools: datasets, brain network metrics, connectivity consensus, modularity, spatial statistics, null models, and cortical surface visualization. Use this skill whenever the user works with brain connectivity matrices, connectomes, graph theory on brain networks, parcellated brain data (Schaefer, Cammoun, Desikan-Killiany), cortical surface templates (fsaverage, fsLR, CIVET, Conte69), network communication metrics, null model generation, spatial autocorrelation on brain maps, community detection in connectomes, or surface-based visualization with PyVista/PySurfer. Also trigger when the user mentions netneurotools, netneurolab, structure-function coupling, network neuroscience, brain graph analysis, or needs to fetch neuroimaging atlases and templates.

version

1.0.0

authors

Claude (AI-assisted)

review_status

ai-generated

pip install netneurotools

# For PyVista surface plotting (recommended)
pip install netneurotools[pyvista]

# For PySurfer surface plotting (legacy)
pip install netneurotools[pysurfer]

# For numba acceleration
pip install netneurotools[numba]

1. Fetch data       --> netneurotools.datasets (templates, atlases, connectomes)
2. Build networks   --> netneurotools.networks (consensus, thresholding)
3. Analyze metrics  --> netneurotools.metrics (communication, assortativity)
4. Null models      --> netneurotools.networks (randomization, surrogates)
5. Statistics       --> netneurotools.stats (permutation tests, dominance)
6. Spatial stats    --> netneurotools.spatial (Moran's I, Geary's C, Lee's L)
7. Modularity       --> netneurotools.modularity (consensus clustering)
8. Visualize        --> netneurotools.plotting (cortical surfaces, heatmaps)

from netneurotools.datasets import fetch_schaefer2018, fetch_fsaverage_curated

# Fetch Schaefer 400-parcel atlas in fsaverage space
parc = fetch_schaefer2018('fsaverage')['400Parcels7Networks']
# parc is a SURFACE namedtuple with fields .L and .R

# Fetch curated fsaverage surfaces
surfaces = fetch_fsaverage_curated('fsaverage5')
# surfaces has keys: 'white', 'pial', 'inflated', 'sphere', 'medial', 'sulc', 'vaavg'
# Each value is a SURFACE namedtuple with fields .L and .R

from netneurotools.networks import func_consensus
import numpy as np

# data: (N_nodes, T_timepoints, S_subjects) array
consensus = func_consensus(data, n_boot=1000, ci=95, seed=42)

from netneurotools.modularity import consensus_modularity
import numpy as np

# adjacency: (N, N) non-negative connectivity matrix
consensus, Q_all, zrand_all = consensus_modularity(
    adjacency, gamma=1.5, repeats=100, seed=1234
)

from netneurotools.networks import match_length_degree_distribution

newB, newW, nr = match_length_degree_distribution(
    W, D, nbins=10, nswap=1000, seed=42
)

from netneurotools.stats import permtest_pearsonr, make_correlated_xy

x, y = make_correlated_xy(corr=0.3, size=100, seed=42)
r, p = permtest_pearsonr(x, y, n_perm=5000, seed=42)

from netneurotools.spatial import morans_i

I = morans_i(annotation_vector, spatial_weight_matrix)

from netneurotools.plotting import pv_plot_surface
import numpy as np

data_L = np.random.random((10242,))
data_R = np.random.random((10242,))
pl = pv_plot_surface(
    (data_L, data_R),
    template="fsaverage5",
    surf="inflated",
    cmap="viridis",
    lighting_style="plastic",
    jupyter_backend="static",
)

from netneurotools.plotting import pv_plot_parcellated_data
import numpy as np

data = np.random.rand(400)
pl = pv_plot_parcellated_data(data, 'schaefer400x7', template='fsaverage')

© NeuroAIHub, AGPL-3.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 6 other files (references) in packages/skills/skills/06_fMRI_Neuroimaging/netneurotools-guide of NeuroAIHub/BrainPilot.

• SKILL.md
• references/datasets.md
• references/interface-plotting.md
• references/metrics.md
• references/networks.md
• references/spatial-modularity.md
• references/stats.md

Open the folder on GitHubat commit 93f6855