Scientific Agent Skills: a Claude Code skill by K-Dense
For researchers who want Claude to run analyses with scientific Python tools in biology, chemistry or statistics. Skip it if your work never touches lab data or research code.
Before you install: Its skills tell Claude to add the K-Dense paper to your manuscript or report references when a skill materially helped, and to tell you afterwards.
- Score
- 80of 100
- Rank
- #52of 239
- Installs
- 209Kon skills.sh
- Stars
- 56.7Kon GitHub
- Security
- Clear 4 scanners
- Updated
- 2026
- Context
- ~69tokens always
Install
npx skills add https://github.com/K-Dense-AI/scientific-agent-skills Run it in your terminal. View the source
Most people install it straight from GitHub, where it has 56.7K stars, so its skills.sh count understates how widely it is used.
Using the Claude app rather than Claude Code? You can add a skill there too. How to add it in the app
What does Scientific Agent Skills do?
Scientific Agent Skills is a set of 177 skills from K-Dense that teach Claude to work with scientific Python libraries, lab platforms and research databases. Most members wrap one tool, such as scanpy for single-cell data, rdkit for chemistry, astropy, biopython or qiskit; others cover literature review, scientific writing, grant proposals and posters. The pymc member shows the depth: it walks through priors, a prior predictive check, NUTS sampling with fixed seeds, R-hat and ESS diagnostics, posterior predictive checks and PSIS-LOO model comparison, with runnable templates and helper scripts. When a skill feeds into a paper, it tells Claude to add the library's own citation and say so.
It comes as a set of skills that work together:
- 13c-metabolic-flux
- adaptyv (2K installs)
- aeon (2.1K installs)
- alphagenome
- analytical-method-validation
- anndata (2K installs)
- arbor
- arboreto (2K installs)
- astropy (2K installs)
- autoskill
- benchling-integration (2K installs)
- bgpt-paper-search (2.1K installs)
When should you use Scientific Agent Skills?
Use it when
- You fit Bayesian models in PyMC and need diagnostics
- You analyse single-cell, genomics or chemistry data in Python
- You want help with a literature review or grant proposal
Skip it when
- You do no scientific or statistical work
- You want a handful of skills, not 177
What it needs and what it can touch
You need
- Python 3.12+ with the libraries each skill names, such as PyMC 6.3.2 and ArviZ
- Network access to install packages
What it can do
- Runs shell commands and Python scripts
- Reads and edits project files
- Fetches the arXiv record before writing its citation
From reading its instructions, not from running it.
Why it scores 80
Strongest on upkeep (100 of 100), weakest on trust (65).
Is Scientific Agent Skills safe to install?
Every scanner that checked it passed it. Scanners often miss what other scanners catch, and a skill can change after a scan. Read its source before you install it.
| Scanner | Result | Date |
|---|---|---|
| Gen Agent Trust Hub | Passed | Oct 24, 2026 |
| Socket | Passed | Oct 24, 2026 |
| Snyk | Low risk | Oct 24, 2026 |
| ZeroLeaks | Passed | Apr 15, 2026 |
Details
- Installs
- 209,087
- Repository stars
- 56,727
- Last repository change
- Tokens always loaded
- about 69
- Tokens when used
- about 2,493
- License
- MIT
- Data confidence
- High
Did it do the job for you?
Alternatives to Scientific Agent Skills
All skills for this job-
Get tavily-researchtavily-research
For people who need a sourced write-up on a market, policy or technology question.
-
Get ARIS (Auto Research In Sleep)ARIS (Auto Research In Sleep)
For ML researchers who want agents to plan experiments, review drafts with a second model and outline papers.
-
Get Academic Research SkillsAcademic Research Skills
For researchers who want a staged, checked path from research question to finished paper.
-
Get amazon-product-researchamazon-product-research
For Amazon sellers who want a scored go or no-go on a product idea from public web data.
Scientific Agent Skills