Scientific Agent Skills
Selected general-purpose data, research and visual tools.
MIT licence
What it does
A broad upstream collection with a selected subset listed here for general data analysis, machine learning, research, documents and image work. Specialist laboratory, biomedical and physical-science workflows from the spreadsheet are excluded from this directory selection.
Installation
- The installer offers the upstream collection, including specialist science skills not listed here. Select only the skills you need; installing instructions does not install their runtime dependencies.
- Individual skills have their own licence metadata, which can differ from the repository MIT licence.
Skills in this collection
45 skills
-
Data
aeon
Analyse and forecast time-series data.
-
Data
arbor
Organise repeatable AI research experiments.
-
Operations
autoskill
Find recurring workflows in Screenpipe history.
-
Writing
citation-management
Find and validate research citations.
-
Operations
consciousness-council
Explore a decision through simulated perspectives.
-
Data
dask
Scale Python data processing with Dask.
-
Data
database-lookup
Retrieve facts from documented database APIs.
-
Data
exa-search
Search and extract web evidence with Exa.
-
Design
generate-image
Generate and edit images through OpenRouter.
-
Data
geopandas
Work with geographic vector data in Python.
-
Operations
get-available-resources
Check usable hardware before heavy workloads.
-
Design
infographics
Create and review data-led infographics.
-
Writing
liteparse
Extract document text with layout information.
-
Writing
literature-review
Build a traceable literature review.
-
Writing
markdown-mermaid-writing
Write Markdown documents and Mermaid diagrams.
-
Marketing
market-research-reports
Prepare evidence-backed market research reports.
-
Writing
markitdown
Convert documents to Markdown with MarkItDown.
-
Data
matlab
Build MATLAB and GNU Octave workflows.
-
Data
matplotlib
Create precisely controlled Matplotlib plots.
-
Data
modal
Run Python workloads and models on Modal.
-
Data
networkx
Analyse and visualise networks with NetworkX.
-
Operations
open-notebook
Organise a self-hosted research notebook.
-
Data
optimize-for-gpu
Measure and improve Python GPU performance.
-
Data
parallel-web
Run web research with Parallel CLI.
-
Data
polars
Build fast data pipelines with Polars.
-
Data
pymc
Build and diagnose Bayesian PyMC models.
-
Data
pymoo
Solve multi-objective problems with pymoo.
-
Data
pytorch-lightning
Structure model training with PyTorch Lightning.
-
Writing
pyzotero
Manage Zotero references with Python.
-
Data
scikit-learn
Build and evaluate scikit-learn pipelines.
-
Data
seaborn
Create readable Seaborn statistical charts.
-
Data
shap
Explain model predictions with SHAP.
-
Data
simpy
Simulate queues and shared resources.
-
Data
stable-baselines3
Train and evaluate reinforcement-learning agents.
-
Data
statistical-analysis
Choose and report appropriate statistical analyses.
-
Data
statistical-power
Plan sample sizes and statistical power.
-
Data
statsmodels
Fit and diagnose statistical models in Python.
-
Data
sympy
Solve exact symbolic mathematics with SymPy.
-
Data
timesfm-forecasting
Forecast time series with TimesFM.
-
Data
torch-geometric
Build graph neural networks with PyTorch.
-
Data
transformers
Use pretrained models with Transformers.
-
Data
umap-learn
Explore dimensionality reduction with UMAP.
-
Data
usfiscaldata
Retrieve US Treasury fiscal data.
-
Data
vaex
Explore large tabular datasets with Vaex.
-
Data
zarr-python
Store chunked arrays with Zarr-Python.