Research Orchestrator
by Tristan578
Coordinates academic research workflow - delegates analysis, correlation, writing, and review tasks to specialist agents
Skill Details
Repository Files
1 file in this skill directory
name: research-orchestrator description: Coordinates academic research workflow - delegates analysis, correlation, writing, and review tasks to specialist agents allowed-tools: [Skill, Task, Read, Write, TodoWrite]
Research Orchestrator
You manage research projects from start to finish. You delegate work to specialist agents using the Task and Skill tools, and ensure quality at each stage.
Your Workflow
Stage 1: Extract Data from Papers
Who does it: Use Skill tool to invoke academic-researcher
Input: PDF files in papers/ folder
Output: results/parsed_papers.json (structured data)
Your job: Verify the JSON file has data for all papers
How to delegate:
Use Skill tool with command: "academic-researcher"
The skill will read PDFs and create parsed_papers.json
Stage 2: Calculate Statistics
Who does it: Use Skill tool to invoke academic-researcher again
Input: results/parsed_papers.json
Output: results/correlation_analysis.json
Your job: Verify correlation coefficients are valid (between -1 and 1)
How to delegate:
Use Skill tool with command: "academic-researcher"
Ask it to calculate correlations from parsed data
Stage 3: Write Article Draft
Who does it: Use Skill tool to invoke technical-copywriter
Input: results/correlation_analysis.json
Output: results/draft_article.md
Your job: Verify article has proper structure and citations
How to delegate:
Use Skill tool with command: "technical-copywriter"
The skill will read analysis results and write article
Stage 4: Review Quality
Who does it: Use Skill tool to invoke research-antagonist
Input: results/draft_article.md
Output: results/review_feedback.json
Your job: Check if approved or needs revision
How to delegate:
Use Skill tool with command: "research-antagonist"
The skill will review the draft and provide feedback
If revision needed, go back to Stage 3 and invoke technical-copywriter again. If approved, workflow complete.
When to Escalate to Human
Stop and ask for human help if:
- Any stage fails 3 times in a row
- Data extraction returns empty results
- Statistical calculations produce impossible values (r > 1, p > 1)
- Review finds critical errors that can't be auto-fixed
Success Criteria
Project complete when:
- All JSON files exist and have valid data
- Draft article is complete with citations
- Antagonist status = "APPROVED"
Related Skills
Dask
Parallel/distributed computing. Scale pandas/NumPy beyond memory, parallel DataFrames/Arrays, multi-file processing, task graphs, for larger-than-RAM datasets and parallel workflows.
Scikit Survival
Comprehensive toolkit for survival analysis and time-to-event modeling in Python using scikit-survival. Use this skill when working with censored survival data, performing time-to-event analysis, fitting Cox models, Random Survival Forests, Gradient Boosting models, or Survival SVMs, evaluating survival predictions with concordance index or Brier score, handling competing risks, or implementing any survival analysis workflow with the scikit-survival library.
Polars
Fast DataFrame library (Apache Arrow). Select, filter, group_by, joins, lazy evaluation, CSV/Parquet I/O, expression API, for high-performance data analysis workflows.
Scikit Survival
Comprehensive toolkit for survival analysis and time-to-event modeling in Python using scikit-survival. Use this skill when working with censored survival data, performing time-to-event analysis, fitting Cox models, Random Survival Forests, Gradient Boosting models, or Survival SVMs, evaluating survival predictions with concordance index or Brier score, handling competing risks, or implementing any survival analysis workflow with the scikit-survival library.
Dask
Distributed computing for larger-than-RAM pandas/NumPy workflows. Use when you need to scale existing pandas/NumPy code beyond memory or across clusters. Best for parallel file processing, distributed ML, integration with existing pandas code. For out-of-core analytics on single machine use vaex; for in-memory speed use polars.
Anndata
Data structure for annotated matrices in single-cell analysis. Use when working with .h5ad files or integrating with the scverse ecosystem. This is the data format skill—for analysis workflows use scanpy; for probabilistic models use scvi-tools; for population-scale queries use cellxgene-census.
Matplotlib
Low-level plotting library for full customization. Use when you need fine-grained control over every plot element, creating novel plot types, or integrating with specific scientific workflows. Export to PNG/PDF/SVG for publication. For quick statistical plots use seaborn; for interactive plots use plotly; for publication-ready multi-panel figures with journal styling, use scientific-visualization.
Dashboard Design
USE THIS SKILL FIRST when user wants to create and design a dashboard, ESPECIALLY Vizro dashboards. This skill enforces a 3-step workflow (requirements, layout, visualization) that must be followed before implementation. For implementation and testing, use the dashboard-build skill after completing Steps 1-3.
Writing Effective Prompts
Structure Claude prompts for clarity and better results using roles, explicit instructions, context, positive framing, and strategic organization. Use when crafting prompts for complex tasks, long documents, tool workflows, or code generation.
Flowchart Creator
Create HTML flowcharts and process diagrams with decision trees, color-coded stages, arrows, and swimlanes. Use when users request flowcharts, process diagrams, workflow visualizations, or decision trees.