> ## Documentation Index
> Fetch the complete documentation index at: https://docs.galagos.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Pre-built Skills

> Galagos ships with 250+ pre-built Skills covering protein structure, single-cell, RNA-seq, variant analysis, chemistry, drug discovery, and clinical workflows. Browse, attach, toggle.

The catalogue ships with **250+ prebuilt Skills** covering the most common bioinformatics tools and workflows. Each Skill encodes how a tool or procedure should be run — parameters, input/output handling, interpretation — so the agent can apply it correctly without manual configuration.

> [!TIP]
>
> You do not need to browse the catalogue before starting an analysis. The agent suggests relevant Skills as you describe your question. Browse when you want to see what's available or pre-attach Skills before running.

Open the catalogue from the wrench icon in a project's right sidebar, then click **+ Add**. Search filters as you type; each card has a one-line description and a click-through for the full text.

## Structural biology and protein design

Predict, model, and analyse protein and macromolecular structures.

| Skill | Description |
| - | - |
| **AlphaFold** | Predict protein structure from amino acid sequence using deep learning. |
| **AlphaFold-Database** | Look up precomputed AlphaFold predictions for known proteins. |
| **ColabFold** | Lighter-weight AlphaFold-compatible structure prediction. |
| **RoseTTAFold** | Three-track neural network for protein structure prediction. |
| **AutoDock Vina** | Molecular docking — predict how small molecules bind protein targets. |

## Single-cell biology

Process and interpret single-cell RNA sequencing data.

| Skill | Description |
| - | - |
| **Scanpy** | Python toolkit for clustering, trajectory inference, and visualisation. |
| **Seurat** | R toolkit with integration, annotation, and multimodal support. |
| **CellRanger** | Align and quantify 10x Genomics single-cell sequencing data. |

## RNA-seq and differential expression

Align, quantify, and run differential expression on bulk RNA-seq.

| Skill | Description |
| - | - |
| **STAR** | Fast splice-aware RNA-seq alignment to a reference genome. |
| **HISAT2** | Graph-based alignment with splice-site awareness. |
| **Bowtie2** | Fast short-read alignment for DNA and RNA. |
| **featureCounts** | Read counting for genes and genomic features. |
| **DESeq2** | Differential expression with negative-binomial modelling. |
| **pydeseq2** | Python implementation of DESeq2 for end-to-end notebooks. |
| **edgeR** | Differential expression with empirical Bayes estimation. |

## Variant analysis and genomics

Call, annotate, and filter genetic variants.

| Skill | Description |
| - | - |
| **GATK HaplotypeCaller** | Call germline SNPs and indels via local de-novo haplotype assembly. |
| **bcftools** | Variant calling, filtering, and manipulation for VCF and BCF. |
| **SnpEff** | Annotate and predict the functional effects of variants. |
| **alphagenome** | Genome-aware Skill for variant interpretation tasks. |

## Sequence analysis and phylogenetics

Search, align, and compare biological sequences and trees.

| Skill | Description |
| - | - |
| **BLAST** | Search sequence databases for regions of local similarity. |
| **HMMER** | Profile hidden Markov models for sensitive homology searches. |
| **Clustal Omega** | Multiple sequence alignment for proteins and nucleotides. |
| **IQ-TREE** | Maximum-likelihood phylogenetic inference with model selection. |

## Chemistry and small molecules

Work with chemical structures, compound libraries, and drug-target interactions.

| Skill | Description |
| - | - |
| **RDKit** | Cheminformatics toolkit — descriptors, similarity, fingerprints. |
| **ChEMBL** | Look up bioactivity and target data for small molecules. |

## Workflow Skills

Multi-step procedures that chain tools together.

| Skill | Description |
| - | - |
| **Standard RNA-seq pipeline** | Raw reads to differential expression: alignment → quantification → DEA. |
| **Literature review** | Structured review and synthesis of recent publications relevant to your question. |
| **Replication study** | Reproduce a published analytical workflow against your own data. |

> [!NOTE]
>
> The examples on this page are a selection of the full catalogue, which includes 250+ Skills across Genomics & Pipelines, Drug Discovery, Single-Cell Biology, Chemistry & Molecules, Protein Structure & Design, Proteomics & Metabolomics, and Clinical & Medical. Open the catalogue inside a project to see the complete list.

## How to use Skills in a project

**1. Open the Skills panel**

Click the wrench icon in the project's right sidebar.

**2. Attach what you need**

Click **+ Add** to open the catalogue. Search or browse, then click **+ Add** on a card to attach it.

**3. Toggle on the active ones**

The panel header shows the active count (e.g. *"3 of 5 enabled"*). Toggle a Skill on; the next prompt picks it up.

**4. Verify in the Steps list**

When the agent consults a Skill it logs a step like *"Enable pydeseq2 Skill"*. Open the Steps list to confirm.

## A few tips

* **Don't over-attach.** Two or three Skills per project is usually enough. Twenty competing Skills produce muddier results, not better ones.
* **Read the description before attaching.** A Skill is a brief — read it to make sure it matches what you want.
* **Use workflow Skills for repeatable pipelines.** A workflow Skill is far more reliable than re-describing the same five-step procedure each time.

## Related pages

> [card] [Skills overview](https://docs.galagos.ai/skills/overview.md)
>
> What Skills are and how the panel works.

> [card] [Custom Skills](https://docs.galagos.ai/skills/custom-skills.md)
>
> Write your own when the catalogue doesn't fit.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.