Skip to content

Commit 8d1ace7

Browse files
authored
Merge pull request #60 from sixiang-svg/update-research-part-3
feat: update research skills part 3
2 parents 4eef9d6 + fb8931a commit 8d1ace7

20 files changed

Lines changed: 342 additions & 2394 deletions

File tree

skills/Research/deep-research/SKILL.md

Lines changed: 8 additions & 10 deletions
Original file line numberDiff line numberDiff line change
@@ -1,16 +1,14 @@
11
---
2-
category: Research
32
id: deep-research
43
name: Deep Research
5-
description: Step-by-step guidance for deep research.
6-
Comprehensive research assistant that synthesizes information from multiple sources with citations.
7-
Use when: conducting in-depth research, gathering sources, writing research summaries, analyzing topics
8-
from multiple perspectives, or when user mentions research, investigation, or needs synthesized analysis
9-
with citations.
10-
license: MIT
11-
metadata:
12-
author: awesome-llm-apps
13-
version: "1.0.0"
4+
description: Comprehensive research assistant that synthesizes information from multiple sources with verified citations and structured analysis.
5+
category: Research
6+
author: awesome-llm-apps
7+
version: "1.0.0"
8+
requires: []
9+
examples:
10+
- Research the current state of solid-state battery technology and provide citations.
11+
- Synthesize the competing theories on dark matter from multiple scientific sources.
1412
---
1513

1614
# Deep Research

skills/Research/deepchem/SKILL.md

Lines changed: 20 additions & 8 deletions
Original file line numberDiff line numberDiff line change
@@ -1,20 +1,32 @@
11
---
2-
category: Research
32
id: deepchem
43
name: Deepchem
5-
description: Molecular machine learning toolkit. Property prediction (ADMET, toxicity), GNNs (GCN, MPNN), MoleculeNet benchmarks, pretrained models, featurization, for drug discovery ML.
4+
description: Molecular machine learning toolkit for ADMET prediction, graph neural networks, and MoleculeNet benchmarks in drug discovery.
5+
category: Research
6+
requires: []
7+
examples:
8+
- Predict the toxicity of this molecule using a DeepChem GNN model.
9+
- Featurize this list of SMILES strings for a property prediction task.
610
---
711

812
# Deepchem
913

1014
Molecular machine learning toolkit. Property prediction (ADMET, toxicity), GNNs (GCN, MPNN), MoleculeNet benchmarks, pretrained models, featurization, for drug discovery ML.
1115

12-
## When to Use
16+
## Instruction
17+
- Identify the specific drug discovery task, such as toxicity estimation or ADMET property prediction.
18+
- Select and apply the correct featurizer, such as `GraphConv` or `CircularFingerprint`, to prepare SMILES for modeling.
19+
- Load relevant benchmarking datasets from MoleculeNet if required for model evaluation.
20+
- Initialize and configure deep learning models, focusing on Graph Neural Networks (GNNs) like GCN or MPNN.
21+
- Perform property predictions using either pre-trained models or custom-trained architectures.
22+
- Evaluate model performance using standard scientific metrics like RMSE or ROC-AUC.
1323

14-
- You need help with deepchem.
15-
- You want a clear, actionable next step.
24+
## When to Use
25+
- When predicting the biological activity or safety profiles of candidate molecules.
26+
- When benchmarking molecular machine learning models against established drug discovery datasets.
27+
- When featurizing chemical structures for integration into machine learning workflows.
1628

1729
## Output
18-
19-
- Summary of goals and plan
20-
- Key tips and precautions
30+
- Property predictions or toxicity scores for the input molecular set.
31+
- Model performance summaries and benchmarking results.
32+
- A structured overview of the featurization and modeling pipeline.

skills/Research/deeptools/SKILL.md

Lines changed: 20 additions & 8 deletions
Original file line numberDiff line numberDiff line change
@@ -1,20 +1,32 @@
11
---
2-
category: Research
32
id: deeptools
43
name: deepTools
5-
description: NGS analysis toolkit. BAM to bigWig conversion, QC (correlation, PCA, fingerprints), heatmaps/profiles (TSS, peaks), for ChIP-seq, RNA-seq, ATAC-seq visualization.
4+
description: Analysis and visualization of NGS data, supporting BAM to bigWig conversion, QC plots, and TSS coverage heatmaps.
5+
category: Research
6+
requires: []
7+
examples:
8+
- Generate a PCA plot for my ChIP-seq samples to check for correlation.
9+
- How do I convert my BAM files to bigWig format for visualization in IGV?
610
---
711

812
# deepTools
913

1014
NGS analysis toolkit. BAM to bigWig conversion, QC (correlation, PCA, fingerprints), heatmaps/profiles (TSS, peaks), for ChIP-seq, RNA-seq, ATAC-seq visualization.
1115

12-
## When to Use
16+
## Instruction
17+
- Analyze the sequencing dataset (ChIP-seq, RNA-seq, etc.) to identify necessary QC or visualization steps.
18+
- Guide the conversion of BAM files to bigWig format to facilitate visualization in genome browsers.
19+
- Perform sample correlation or PCA analysis to assess biological replicate consistency.
20+
- Create heatmaps or enrichment profiles centered on genomic features like TSS or peak regions.
21+
- Apply appropriate normalization methods, such as RPKM or 1x coverage, to ensure inter-sample comparability.
22+
- Recommend plotting parameters to optimize the visual clarity of sequencing signal enrichment.
1323

14-
- You need help with deeptools.
15-
- You want a clear, actionable next step.
24+
## When to Use
25+
- When visualizing the genomic distribution of NGS reads across different experimental conditions.
26+
- When detecting sample outliers or batch effects through statistical quality control.
27+
- When generating high-quality figures for genomic signal profiles and peak heatmaps.
1628

1729
## Output
18-
19-
- Summary of goals and plan
20-
- Key tips and precautions
30+
- Technical steps and commands for BAM processing, normalization, and track generation.
31+
- Interpretation of QC plots and their impact on the overall data reliability.
32+
- Visual plans for generating publication-ready genomic enrichment figures.

skills/Research/denario/SKILL.md

Lines changed: 29 additions & 137 deletions
Original file line numberDiff line numberDiff line change
@@ -1,8 +1,12 @@
11
---
2-
category: Research
32
id: denario
43
name: Denario
5-
description: Guidance and answers for denario.
4+
description: Multiagent AI system automating the scientific research pipeline from hypothesis generation to publication-ready LaTeX manuscripts.
5+
category: Research
6+
requires: []
7+
examples:
8+
- Generate a novel research hypothesis for this climate dataset using Denario.
9+
- Write a journal-formatted LaTeX manuscript based on my recent computational results.
610
---
711

812
# Denario
@@ -11,6 +15,14 @@ description: Guidance and answers for denario.
1115

1216
Denario is a multiagent AI system designed to automate scientific research workflows from initial data analysis through publication-ready manuscripts. Built on AG2 and LangGraph frameworks, it orchestrates multiple specialized agents to handle hypothesis generation, methodology development, computational analysis, and paper writing.
1317

18+
## Instruction
19+
- Initialize the research project by defining the data context and available computational tools.
20+
- Orchestrate specialized agents to generate novel research hypotheses based on the data description.
21+
- Develop a structured methodology that outlines the computational experiments needed to test the hypothesis.
22+
- Execute the methodology to generate computational results, analysis, and visualizations.
23+
- Synthesize all findings into a journal-formatted LaTeX manuscript using specific styles like `Journal.APS`.
24+
- Ensure research reproducibility by maintaining structured outputs for every stage of the pipeline.
25+
1426
## When to Use This Skill
1527

1628
Use this skill when:
@@ -21,22 +33,10 @@ Use this skill when:
2133
- Writing journal-formatted LaTeX papers from research results
2234
- Automating the complete research pipeline from data to publication
2335

24-
## Installation
25-
26-
Install denario using uv (recommended):
27-
28-
```bash
29-
uv init
30-
uv add "denario[app]"
31-
```
32-
33-
Or using pip:
34-
35-
```bash
36-
uv pip install "denario[app]"
37-
```
38-
39-
For Docker deployment or building from source, see `references/installation.md`.
36+
## Output
37+
- A defined research hypothesis and a detailed methodology document.
38+
- Computational findings and visualizations ready for scientific interpretation.
39+
- A complete LaTeX source package and a formatted PDF of the final manuscript.
4040

4141
## LLM API Configuration
4242

@@ -45,147 +45,47 @@ Denario requires API keys from supported LLM providers. Supported providers incl
4545
- OpenAI
4646
- Other LLM services compatible with AG2/LangGraph
4747

48-
Store API keys securely using environment variables or `.env` files. For detailed configuration instructions including Vertex AI setup, see `references/llm_configuration.md`.
49-
48+
Store API keys securely using environment variables or `.env` files.
5049
## Core Research Workflow
5150

52-
Denario follows a structured four-stage research pipeline:
51+
Denario follows a structured four-stage research pipeline.
5352

5453
### 1. Data Description
5554

56-
Define the research context by specifying available data and tools:
55+
Define the research context by specifying available data and tools
5756

58-
```python
59-
from denario import Denario
60-
61-
den = Denario(project_dir="./my_research")
62-
den.set_data_description("""
63-
Available datasets: time-series data on X and Y
64-
Tools: pandas, sklearn, matplotlib
65-
Research domain: [specify domain]
66-
""")
67-
```
6857

6958
### 2. Idea Generation
7059

71-
Generate research hypotheses from the data description:
60+
Generate research hypotheses from the data description.
7261

73-
```python
74-
den.get_idea()
75-
```
7662

77-
This produces a research question or hypothesis based on the described data. Alternatively, provide a custom idea:
63+
This produces a research question or hypothesis based on the described data. Alternatively, provide a custom idea
7864

79-
```python
80-
den.set_idea("Custom research hypothesis")
81-
```
8265

8366
### 3. Methodology Development
8467

85-
Develop the research methodology:
68+
Develop the research methodology
8669

87-
```python
88-
den.get_method()
89-
```
9070

91-
This creates a structured approach for investigating the hypothesis. Can also accept markdown files with custom methodologies:
71+
This creates a structured approach for investigating the hypothesis. Can also accept markdown files with custom methodologies
9272

93-
```python
94-
den.set_method("path/to/methodology.md")
95-
```
9673

9774
### 4. Results Generation
9875

99-
Execute computational experiments and generate analysis:
76+
Execute computational experiments and generate analysis
10077

101-
```python
102-
den.get_results()
103-
```
10478

105-
This runs the methodology, performs computations, creates visualizations, and produces findings. Can also provide pre-computed results:
79+
This runs the methodology, performs computations, creates visualizations, and produces findings. Can also provide pre-computed results
10680

107-
```python
108-
den.set_results("path/to/results.md")
109-
```
11081

11182
### 5. Paper Generation
11283

113-
Create a publication-ready LaTeX paper:
114-
115-
```python
116-
from denario import Journal
84+
Create a publication-ready LaTeX paper
11785

118-
den.get_paper(journal=Journal.APS)
119-
```
12086

12187
The generated paper includes proper formatting for the specified journal, integrated figures, and complete LaTeX source.
12288

123-
## Available Journals
124-
125-
Denario supports multiple journal formatting styles:
126-
- `Journal.APS` - American Physical Society format
127-
- Additional journals may be available; check `references/research_pipeline.md` for the complete list
128-
129-
## Launching the GUI
130-
131-
Run the graphical user interface:
132-
133-
```bash
134-
denario run
135-
```
136-
137-
This launches a web-based interface for interactive research workflow management.
138-
139-
## Common Workflows
140-
141-
### End-to-End Research Pipeline
142-
143-
```python
144-
from denario import Denario, Journal
145-
146-
# Initialize project
147-
den = Denario(project_dir="./research_project")
148-
149-
# Define research context
150-
den.set_data_description("""
151-
Dataset: Time-series measurements of [phenomenon]
152-
Available tools: pandas, sklearn, scipy
153-
Research goal: Investigate [research question]
154-
""")
155-
156-
# Generate research idea
157-
den.get_idea()
158-
159-
# Develop methodology
160-
den.get_method()
161-
162-
# Execute analysis
163-
den.get_results()
164-
165-
# Create publication
166-
den.get_paper(journal=Journal.APS)
167-
```
168-
169-
### Hybrid Workflow (Custom + Automated)
170-
171-
```python
172-
# Provide custom research idea
173-
den.set_idea("Investigate the correlation between X and Y using time-series analysis")
174-
175-
# Auto-generate methodology
176-
den.get_method()
177-
178-
# Auto-generate results
179-
den.get_results()
180-
181-
# Generate paper
182-
den.get_paper(journal=Journal.APS)
183-
```
184-
185-
### Literature Search Integration
186-
187-
For literature search functionality and additional workflow examples, see `references/examples.md`.
188-
18989
## Advanced Features
19090

19191
- **Multiagent orchestration**: AG2 and LangGraph coordinate specialized agents for different research tasks
@@ -194,18 +94,10 @@ For literature search functionality and additional workflow examples, see `refer
19494
- **Flexible input**: Manual or automated at each pipeline stage
19595
- **Docker deployment**: Containerized environment with LaTeX and all dependencies
19696

197-
## Detailed References
198-
199-
For comprehensive documentation:
200-
- **Installation options**: `references/installation.md`
201-
- **LLM configuration**: `references/llm_configuration.md`
202-
- **Complete API reference**: `references/research_pipeline.md`
203-
- **Example workflows**: `references/examples.md`
204-
20597
## Troubleshooting
20698

20799
Common issues and solutions:
208-
- **API key errors**: Ensure environment variables are set correctly (see `references/llm_configuration.md`)
100+
- **API key errors**: Ensure environment variables are set correctly
209101
- **LaTeX compilation**: Install TeX distribution or use Docker image with pre-installed LaTeX
210102
- **Package conflicts**: Use virtual environments or Docker for isolation
211103
- **Python version**: Requires Python 3.12 or higher

0 commit comments

Comments
 (0)