Skip to main content

Getting started

caution

Seqera AI is currently in beta. Features and commands may change as we continue to improve the product.

This guide walks you through installing Seqera AI and running your first commands.

Prerequisites​

Before you begin, ensure you have:

Step 1: Install the CLI​

Install Seqera AI using pip:

pip install seqera-ai

Verify the installation:

seqera --version

Step 2: Login to Seqera Platform​

Authenticate with your Seqera Platform account:

seqera login

This opens your browser to complete authentication. Once successful, you'll see a confirmation message in your terminal.

Step 3: Start the AI assistant​

Launch the interactive assistant:

seqera ai

You'll see a welcome message and prompt where you can start typing commands in natural language.

Working with Nextflow​

Start from your Nextflow project directory to get the most out of Seqera AI:

cd /path/to/your/pipeline
seqera ai

Understand your pipeline​

> Show me the structure of main.nf
> What processes are defined in this pipeline?

Generate a configuration file​

> /config

The assistant will analyze your project and create an appropriate nextflow.config file.

Debug your pipeline​

> /debug

Run nextflow lint and preview to check your pipeline for errors.

> Why is my pipeline failing?

The assistant analyzes your code and provides insights into what might be wrong.

Generate a schema​

> /schema

Create a nextflow_schema.json file for your pipeline parameters.

Convert scripts to Nextflow​

> /convert-python-script

Convert a Python script in your working directory to a Nextflow process.

Working with Seqera Platform​

Once you're ready to run pipelines at scale, use Seqera Platform:

List your workflows​

> List my recent workflows

The assistant will show your recent workflow runs from Seqera Platform.

Launch a pipeline​

> Launch the nf-core/rnaseq pipeline with the test profile

The assistant will guide you through launching a workflow, asking for any required parameters.

Debug a failed run​

> Why did my last workflow fail?
> Get the logs for the failed task in my last run

The assistant analyzes logs and provides insights into what went wrong and how to fix it.

Working with nf-core modules​

Seqera AI provides access to over 1,000 nf-core modules for common bioinformatics tasks:

Search for modules​

> Find nf-core modules for sequence alignment
> What modules are available for variant calling?

Get module details​

> Show me how to use the nf-core/bwa/mem module

The assistant returns detailed information including input/output schemas and ready-to-run Nextflow commands.

Run a module​

> Run FastQC on my FASTQ files

The assistant can generate the exact Nextflow command with proper parameters for your data.

Working with data​

Seqera AI helps you manage data through Platform data links and access reference datasets:

> List my data links
> Show me the contents of my S3 data link

Download and upload files​

> Generate a download URL for results/final_report.html
> Upload my local results to the data link

Access reference data​

> Find the human reference genome GRCh38
> Search for RNA-Seq test data

Building containers with Wave​

Seqera AI can create containerized environments using Wave, without requiring you to write Dockerfiles:

Create a container with conda packages​

> Create a container with samtools and bwa from bioconda

Create a container with pip packages​

> Build a container with pandas, numpy, and scikit-learn

Get a container for a specific tool​

> I need a container with FastQC version 0.12.1

The assistant will generate a Wave container URL that you can use directly in your Nextflow pipelines or pull with Docker.

Working with local files​

Seqera AI can interact with files in your current working directory. Start the assistant from your project folder:

cd /path/to/your/project
seqera ai

Then ask the assistant to help with local tasks:

> Show me the structure of main.nf
> Add a new process to handle quality control
note

Local file operations are controlled by approval modes. By default, the assistant will ask for your approval before making changes outside your working directory or running potentially dangerous commands.

Using slash commands​

Seqera AI includes built-in slash commands for common workflows. Type / to see all available commands:

CommandDescription
/configGenerate a nextflow.config file
/schemaGenerate a Nextflow schema
/debugRun nextflow lint and preview
/debug-last-runDebug the last local run
/debug-last-run-on-seqeraDebug the last Platform run
/migrate-from-wdlConvert WDL to Nextflow
/migrate-from-snakemakeConvert Snakemake to Nextflow
/convert-python-scriptConvert Python script to Nextflow
/convert-r-scriptConvert R script to Nextflow
/convert-jupyter-notebookConvert Jupyter notebook to Nextflow
/write-nf-testWrite nf-tests for your pipeline

Command-line options​

Customize your session with these options:

# Start in a specific directory
seqera ai -w /path/to/project

# Set approval mode for local commands
seqera ai -a full

Exiting the assistant​

To end your session:

  • Type exit or quit
  • Press Ctrl+C

Your conversation history is preserved for the session but not stored permanently.

Next steps​