Scientific skills for AI agents
Ordered by install count from the source, then by repository stars on GitHub. Only cards with a checked description are included.
-
1
474,500installs
repo-intake-and-plan
Scans a repository through its README and project files, then builds a small reproduction plan. Use it before running experiments.
skill · lllllllama · MIT
-
2
474,400installs
minimal-run-and-audit
Captures and normalizes execution evidence from deep learning smoke tests and inference commands into standardized repro_outputs/ files.
skill · lllllllama · MIT
-
3
335,600installs
ai-research-explore
Governs deep learning research exploration on top of an existing baseline. It isolates candidate experiments and generates structured comparison reports.
skill · lllllllama · MIT
-
4
335,300installs
ai-research-reproduction
This skill guides a repository-grounded reproduction run from its README to an evidence bundle. Use it to test a documented model run without quietly changing the experiment.
skill · lllllllama · MIT
-
5
335,000installs
explore-code
A skill for cautious code experiments in deep learning repositories: it transplants modules and adapts architectures on an isolated branch, with rollback-aware records.
skill · lllllllama · MIT
-
6
334,700installs
explore-run
Plans bounded exploratory machine learning runs with comparability caveats and saves summaries to explore_outputs.
skill · lllllllama · MIT
-
7
334,700installs
run-train
Runs a selected training command and organizes the resulting evidence. Use it for startup checks, short-run verification, full training kickoff, or resume handling.
skill · lllllllama · MIT
-
8
38,500installs
anysearch
This skill gives an agent web search, domain-specific queries, parallel batch search, and page extraction. Use it when a task needs current information or a claim needs checking.
skill · anysearch-ai · Apache-2.0
-
9
18,700installs
nature-figure
A skill that builds, revises and audits scientific figures for manuscripts in Python or R. Useful when plots go to submission, not to a report.
skill · Yuan1z0825 · Apache-2.0
-
10
17,400installs
nature-polishing
A skill that polishes, translates and tightens existing academic prose. Facts, terminology and evidence boundaries stay intact.
skill · Yuan1z0825 · Apache-2.0
-
11
15,600installs
nature-citation
Finds Nature and CNS-family papers for manuscript claims, maps claims to sources, and prepares a reference-manager export. Use it when the journal scope matters.
skill · Yuan1z skill, refactored into static/dynamic layers · Apache-2.0
-
12
15,300installs
nature-data
nature-data drafts or audits manuscript data and code availability statements, dataset access routes, repository plans, and FAIR metadata.
skill · Yuan1z skill, refactored into static/dynamic layers · Apache-2.0
-
13
15,200installs
nature-academic-search
Searches research literature, checks citations, and builds MeSH strategies. Useful for source discovery, bibliography work, and citation audits.
skill · Yuan1z0825 · Apache-2.0
-
14
14,900installs
nature-reader
Creates source-grounded Chinese-English paper readers with aligned text, tables, and equations. Handles full translations, excerpt parsing, and targeted questions.
skill · Community contribution, refactored into static/dynamic layers · Apache-2.0
-
15
13,800installs
nature-reviewer
The skill runs a Nature-style referee assessment of a scientific manuscript: three independent reviewer reports plus a post-review synthesis. Useful before you submit.
skill · Yuan1z0825 · Apache-2.0
-
16
13,100installs
learn
A six-phase research workflow that turns source material into notes, an outline, or a publish-ready draft. Use it for unfamiliar domains and multi-source work.
skill · tw93 · MIT
-
17
11,800installs
nature-literature-pipeline
Automates academic literature discovery, six-dimension paper scoring, and digest delivery to chat apps. Useful for running daily research scans via cron.
skill · Jiahao8595 · MIT
-
18
11,800installs
nature-statistics
Audits and revises statistical reporting in research manuscripts, from experimental units to figure statistics. Use it to review a Methods section, address reviewer comments, or align figures with results.
skill · Yuan1z0825 · Apache-2.0
-
19
10,800installs
academic-paper
This skill runs a 12-agent academic pipeline across 11 modes. It guides projects from literature searches and outlines to citation audits and rebuttal reviews.
skill · Imbad0202 · LicenseRef-Unknown
-
20
10,400installs
nature-paper-card
Turns one scientific paper into a structured deep-reading card: methods, experiment-to-claim evidence, limitations. Useful when you need the paper taken apart, not summarised.
skill · Yuan1z0825 · Apache-2.0
-
21
9,800installs
academic-paper-reviewer
This skill reviews an academic paper through five role-based perspectives. Use it for peer review, revision checks, or a focused methodology assessment.
skill · Imbad0202 · LicenseRef-Unknown
-
22
7,200installs
academic-pipeline
Coordinates academic research from source gathering and drafting through integrity checks, peer review, revisions, and finalization. Useful for a complete paper workflow.
skill · Imbad0202 · LicenseRef-Unknown
-
23
5,600installs
jupyter-notebook official
Scaffolds reproducible Jupyter notebooks for experiments and tutorials using bundled templates and a helper script. It prevents manual raw JSON editing errors.
skill · openai · Apache-2.0
-
24
4,900installs
start official
Checks connected MCP servers and surveys bio-research tools. Helps researchers get oriented with the plugin before starting a project.
skill · anthropics · Apache-2.0
-
25
4,600installs
statistical-analysis official
Provides statistical guidelines for descriptive stats and anomaly detection. Useful when evaluating distributions and business metrics.
skill · anthropics · Apache-2.0
-
26
3,000installs
consulting-analysis
This skill plans a research framework, then turns collected material into a structured report. Use it for market, financial, brand, or competitor analysis.
skill · bytedance · MIT
-
27
2,800installs
read-arxiv-paper
This skill reads an arXiv paper from its source files and writes a Markdown report. It is useful when research needs to be connected to nanochat.
skill · karpathy · MIT
-
28
2,700installs
academic-paper-review
A skill for automated peer review of academic papers following top-conference standards. Assesses methodology, evaluates novelty, identifies weaknesses, and delivers evidence-based critique.
skill · bytedance · MIT
-
29
2,500installs
systematic-literature-review
A skill to automatically search and synthesize multiple papers by topic from arXiv. Collects research, extracts key findings, and outputs reports in APA, IEEE, or BibTeX format.
skill · bytedance · MIT
-
30
2,400installs
instrument-data-to-allotrope official
Converts laboratory instrument output files (PDF, CSV, Excel, TXT) to Allotrope Simple Model (ASM) JSON or flattened CSV. Auto-detects instrument type and outputs ASM JSON, tabular CSV, and exportable parser code.
skill · anthropics · Apache-2.0
Other collections
- Best Claude Code skills 1,975
- Best Claude Code plugins 7
- Best Claude Code subagents 167
- Best Codex skills 1,733
- Frontend skills 400
- Testing skills 353
- Security skills and rules 310
- SEO and marketing skills 161
- Design skills 297
- DevOps skills 508
- Claude Code design skills 219
- Code review skills 377
- AI agent tools for game development 15
- Marketing skills and agents 48
- UI/UX skills 5
- Writing skills 12
- Claude Code slash commands 7
- GitHub Copilot agent skills 1,603
- Gemini CLI extensions and skills 1,338
- Office skills: PDF, Excel, PowerPoint 144
- Data analysis skills 283
- Research skills 44