Skip to content
Get packs

Guide · Agent skill

How to use the gget bioinformatics agent skill

Updated · written with AI from the project's own documentation and checked against our catalogue data

The gget skill teaches an AI agent to query public bioinformatics databases through gget, a command-line tool and Python package with one consistent interface to 20+ genomics resources. This guide explains what the skill does, how to install it from the ZIP on this page, how to ask for it in plain language, and what it needs to run safely. By the end you will know how to install it, trigger it, and check its results.

gget Bioinformatics QueriesK-Dense-AI/scientific-agent-skills · MIT

Quick start

  1. 1Download and unzip the skill. Download the ZIP from this page and unzip it into ~/.claude/skills/ or into your project's .claude/skills/ folder. Keep the folder structure unchanged.
  2. 2Or upload it in the Claude apps. In the Claude apps, open Settings, go to Capabilities, then Skills, and upload the skill.
  3. 3Create a Python 3.12+ environment. Create a clean virtual environment with Python 3.13 if you want optional Census dependencies, or any Python 3.12 or newer otherwise.
  4. 4Install the pinned gget version. Install gget 0.30.8 with uv pip install and confirm that import gget works.
  5. 5Ask a natural question. Describe the gene, sequence, or list you want to query and the output you need. The agent selects the matching gget module.
  6. 6Save and verify results. Ask for output to be saved, check species and database release, and review any warnings before interpreting the results.

What the gget skill makes the agent do

An agent skill is a folder of instructions and reference files that an AI agent reads when a task matches the skill's description. This one covers gget, which can look up gene information, run BLAST or BLAT searches, download viral sequences, fetch PDB and mmCIF protein structures, retrieve G2P residue annotations, run enrichment analysis, and query OpenTargets, COSMIC, CELLxGENE and 8cube mouse specificity data.

Without the skill, an agent may guess at database formats or endpoints. With it, the agent is directed to use gget's documented modules and to check syntax before querying. The skill is aimed at interactive exploration and simple queries.

  • Gene lookups with gget search and gget info
  • Sequence alignment with gget blast, gget blat and gget diamond
  • Protein structures with gget pdb and residue annotations with gget g2p
  • Enrichment analysis, disease and drug associations, and single-cell queries

Install the gget skill from the ZIP

Download the skill ZIP from this page. You can install it in one of two ways. For a personal setup, unzip it into ~/.claude/skills/. For a single project, unzip it into that project's .claude/skills/ folder. Alternatively, in the Claude apps, open Settings, go to Capabilities, and upload the skill under Skills.

The skill folder contains SKILL.md, a references folder with module catalogue and workflow notes, and a scripts folder with three example Python scripts. Keep the folder structure intact so the agent can find the reference files.

  • Personal install: unzip into ~/.claude/skills/
  • Project install: unzip into the project's .claude/skills/
  • Claude apps: Settings, then Capabilities, then Skills, then upload

Set up the gget Python environment

The skill pins gget 0.30.8 and needs Python 3.12 or newer. The source recommends Python 3.13 if you want the optional Census dependencies, and suggests a clean virtual environment to avoid conflicts. The skill's own install example uses uv.

The skill also notes that most functions query remote servers, so the machine needs network access. Some modules need a one-time gget setup before first use.

  • Python 3.12 or newer; 3.13 is recommended for optional Census dependencies
  • Network access to remote databases
  • Optional: gget setup for the elm, cellxgene and cbio modules
  • Optional: COSMIC account credentials for cosmic downloads
  • Optional: compatible MUSCLE and DIAMOND binaries with OpenMP libraries for local alignment
uv venv --python 3.13 .venv source .venv/bin/activate uv pip install "gget==0.30.8" python -c "import gget"

Ask for it with natural requests

You do not need special syntax. Describe the biological question and mention the database or task if you know it. The agent chooses the matching gget module. Good requests name a gene, a protein, a sequence, a species, or a gene list, and say what you want back, such as a table, a file, or a summary.

Be specific about species and identifiers. The skill warns that gene symbols in cellxgene are case-sensitive, and that gget search matches descriptions and synonyms, so an exact gene_name should be selected rather than the first hit.

  • Look up the gene TP53 in human and summarise its transcripts.
  • Run a BLAST search for this protein sequence and list the top hits.
  • Download SARS-CoV-2 sequences matching these filters and save them to a folder.
  • Run an enrichment analysis on this list of 200 human genes using the Enrichr library and report adjusted p-values.
  • Find PDB structures for this UniProt accession using the mmCIF format.

Check the agent's work

Ask the agent to save results with -o or --out so the query can be reproduced. The skill recommends recording the full library name and release for enrichment, reporting mapped and unmapped counts, and preserving the Ensembl index when saving CSV output.

Open Targets filters are applied after the limit, and expression queries fetch only the first page. Treat results as a sample rather than a complete list unless the source documents otherwise.

  • Save outputs with -o/--out
  • Use -q/--quiet in scripts to reduce progress output
  • Record database release, species, assembly and retrieval date
  • Report mapped and unmapped query and background counts for enrichment
gget info --help gget pdb --help

Safety notes: scripts, network and credentials

This skill runs scripts and uses the network. Its bundled scripts are batch_sequence_analysis.py, enrichment_pipeline.py and gene_analysis.py. Read a script before running it on sensitive data, and check which remote servers it contacts.

COSMIC downloads need an account. The skill says to keep COSMIC and OpenAI credentials in named environment variables or interactive prompts, and not to write real credentials into scripts, notebooks or logs. Results come from live public databases that change over time and are not validated for clinical or diagnostic use.

  • Keep credentials in environment variables or prompts, never in files you share
  • Do not use outputs for clinical or diagnostic decisions
  • Pin gget==0.30.8 for reproducible work

What people use it for

  • A bioinformatics student looks up a gene's transcripts and exons during interactive exploration.
  • A researcher runs BLAST or BLAT on a short sequence to see which known sequences match it.
  • An analyst runs an Enrichr enrichment analysis on a gene list and saves adjusted p-values for review.
  • A developer writes a small scripted lookup that pulls protein structure metadata from PDB for a list of accessions.

Frequently asked questions

How do I install the gget agent skill?

Download the skill ZIP from this page and unzip it into ~/.claude/skills/ for personal use or into a project's .claude/skills/ folder. In the Claude apps you can instead upload it under Settings, then Capabilities, then Skills.

What do I need to run gget?

You need Python 3.12 or newer, network access to remote databases, and the pinned gget 0.30.8 package. Some modules also need gget setup, COSMIC credentials for cosmic downloads, or compatible MUSCLE and DIAMOND binaries for local alignment.

Is gget free to use?

The skill's SKILL.md lists the licence as BSD-2-Clause, while this directory lists it as MIT. Check the LICENSE file in the ZIP for the authoritative terms. Most public queries need no key, but COSMIC downloads require an account.

What is the difference between gget and biopython or bioservices?

The skill says gget is for interactive exploration and simple queries. For batch processing or advanced BLAST it recommends biopython, and for multi-database Python workflows it recommends bioservices.

Why does a gget query fail or return unexpected results?

Public databases change their structure over time, which can break adapters. The skill advises checking upstream release notes, pinning the reviewed version gget==0.30.8, and running gget setup for modules that need it. Check the module help with gget module --help first.

Can I use gget results for clinical decisions?

No. The skill's source states that results come from live public databases that change over time and are not validated for medical use, so they should not inform clinical or diagnostic decisions.

New agent skills, repos and AI tools — every week

The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.

Our verdict

This skill suits bioinformatics students and researchers doing interactive gene and protein exploration, developers who need simple scripted lookups, and analysts running quick enrichment analyses with recorded settings. It is not suited to large batch or high-throughput pipelines, where the skill itself recommends biopython, or to clinical decisions. It also needs network access and, for some features, accounts or extra setup.

Download the skill