Skills denario

🔬

denario

Name: denario
Author: K-Dense-AI

Low Risk ⚙️ External commands📁 Filesystem access🌐 Network access🔑 Env variables

Automate scientific research workflows

Also available from: davila7

Transform datasets into publication-ready research by automating hypothesis generation, methodology development, computational experiments, and LaTeX paper writing. Multiagent AI coordinates specialized agents for end-to-end research pipelines.

Supports: Claude Codex Code(CC)

📊 70 Adequate

Download the skill ZIP

Upload in Claude

Go to Settings → Capabilities → Skills → Upload skill

Toggle on and start using

Test it

Using "denario". Generate research on climate data

Expected outcome:

Research hypothesis: Quantify global warming acceleration using linear regression
Methodology: Time-series decomposition, trend analysis, statistical significance testing
Results: 0.18°C per decade warming trend, p < 0.001
Paper: Complete LaTeX document with figures and bibliography

Using "denario". Analyze customer churn dataset

Expected outcome:

Research idea: Ensemble model combining XGBoost and Random Forest with SMOTE
Methodology: Train/test split, cross-validation, hyperparameter tuning, SHAP analysis
Results: 85% AUC-ROC, key churn factors identified
Paper: Journal-formatted manuscript with performance visualizations

Security Audit

Low Risk

v4 • 1/17/2026

All 369 static findings are FALSE POSITIVES. The skill is documentation-only with bash command examples, API key configuration patterns, and file operations for research project management - all legitimate documented functionality for a scientific research automation tool.

Files scanned

3,198

Lines analyzed

findings

Total audits

Risk Factors

⚙️ External commands (224)

📁 Filesystem access (8)

references/examples.md:467 references/examples.md:14 references/llm_configuration.md:73 references/llm_configuration.md:73 references/llm_configuration.md:73 references/llm_configuration.md:73 references/llm_configuration.md:73 references/llm_configuration.md:73

🌐 Network access (9)

references/installation.md:43 references/installation.md:69 references/installation.md:135 references/installation.md:135 references/llm_configuration.md:29 references/llm_configuration.md:50 references/llm_configuration.md:137 references/llm_configuration.md:202 skill-report.json:6

🔑 Env variables (21)

Audited by: claude View Audit History →

Quality Score

Architecture

100

Maintainability

Content

Community

Security

Spec Compliance

What You Can Build

Accelerate publication pipeline

Automate the journey from raw dataset to formatted journal submission, reducing time spent on methodology design and paper writing.

Structured exploration workflows

Transform exploratory data analysis into reproducible research projects with generated hypotheses, methods, and documentation.

Learn research best practices

Follow AI-generated methodologies to understand structured approaches to computational research and scientific writing.

Try These Prompts

Climate analysis

Analyze global temperature anomaly data from 1880-2023 using pandas, scipy, and matplotlib. Generate research questions about long-term warming trends, develop a methodology with linear regression, and create publication-ready paper.

Gene expression study

Process gene expression microarray data from 500 breast cancer patients with 20000 features. Generate hypotheses about treatment response predictors, create methodology with machine learning classification, and produce LaTeX paper.

Financial forecasting

Analyze monthly unemployment rates from 1950-2023 with GDP and inflation indicators. Generate SARIMAX forecasting methodology, execute analysis with confidence intervals, and write research paper.

Full research pipeline

Use denario to run complete research pipeline: set data description for my IoT sensor dataset, generate research idea for anomaly detection, develop methodology with sklearn, execute analysis, and generate APS-formatted LaTeX paper.

Best Practices

Provide detailed data descriptions including format, size, known challenges, and available tools for better AI-generated outputs
Review and refine generated ideas and methodologies before executing computationally expensive analysis stages
Use version control by committing after each pipeline stage to track research evolution and enable reproducibility

Avoid

Providing vague data descriptions leads to generic, unusable research ideas and methodologies
Skipping methodology review before execution may result in inappropriate statistical approaches for your data
Running full pipeline without intermediate validation wastes compute on flawed research directions

Frequently Asked Questions

What LLM providers does denario support?

OpenAI GPT-4/3.5, Google Vertex AI Gemini/PaLM, and any provider compatible with AG2 or LangGraph frameworks.

Is LaTeX required to use this skill?

LaTeX is only needed for PDF paper generation. You can use the full workflow without it and get LaTeX source files.

Can I use my own research methodology?

Yes, use set_method() with your markdown methodology or file path to bypass automated generation.

What Python version is required?

Python 3.12 or higher is required for this skill to function properly.

Can I provide pre-computed results?

Yes, use set_results() to load existing analysis results and skip the computational execution stage.

What journals are supported for paper formatting?

American Physical Society (APS) format is included. Check documentation for additional journal formats.

Developer Details

Author

K-Dense-AI

License

GPL-3.0 license

Repository

https://github.com/K-Dense-AI/claude-scientific-skills/tree/main/scientific-skills/denario

Ref

main

File structure

📁 references/

📄 examples.md

📄 installation.md

📄 llm_configuration.md

📄 research_pipeline.md

📄 SKILL.md