Guide · Agent skill
How to use the Biopython Bioinformatics Guide agent skill
Updated · written with AI from the project's own documentation and checked against our catalogue data
The biology-biopython skill is a reference guide that teaches an AI agent standard Biopython patterns for sequence work, file parsing, BLAST, NCBI lookups, phylogenetic trees and PDB structures. It is for researchers, students and bioinformatics developers who want an agent to write correct Biopython code. After this guide you will know how to install the skill, phrase requests so it triggers, and what you still need to supply yourself.
Biopython Bioinformatics Guideaiming-lab/AutoResearchClaw · MIT Download the skill Quick start
- 1Download the ZIP. Download the skill ZIP from this page. It contains SKILL.md.
- 2Unzip into a skills folder. Extract it into ~/.claude/skills/ for personal use, or into a project's .claude/skills/ folder.
- 3Or upload in the Claude apps. In the Claude apps, open Settings, go to Capabilities, and upload the skill under Skills.
- 4Install Biopython. Install the Biopython library in your Python environment, since the skill does not include it.
- 5Set your NCBI email. Set Entrez.email in your code before any NCBI request, as the skill requires.
- 6Ask with a natural request. Describe your file type, task and output, for example converting a GenBank file to FASTA.
- 7Review and run the code. Read the generated script, then run it yourself, since the skill itself runs no scripts.
What the skill makes the agent do
The skill does not run analyses on its own. It gives the agent a set of best practices and code snippets for the Biopython library, so that when a request involves DNA, RNA or protein sequences, FASTA or GenBank files, BLAST searches, NCBI database lookups, phylogenetic trees or PDB protein structures, the agent writes Biopython code that follows those patterns.
- Sequence operations: complement, reverse complement, transcription, translation, GC content and molecular weight
- File parsing with SeqIO: reading, writing and converting FASTA and GenBank files
- BLAST: online queries with NCBIWWW, parsing results with NCBIXML, and notes on local BLAST
- NCBI Entrez: searching and fetching records, with the email and rate-limit rules built in
- Phylogenetics with Bio.Phylo: reading Newick, Nexus and PhyloXML trees, drawing and measuring them
- Structure analysis with Bio.PDB: parsing PDB and mmCIF files and walking the Structure > Model > Chain > Residue > Atom hierarchy
Install the skill
Option 1: download the ZIP from this page and unzip it into your personal skills folder, ~/.claude/skills/. Option 2: unzip it into a project's .claude/skills/ folder so it applies only to that project. Option 3: in the Claude apps, open Settings, go to Capabilities, and upload the skill under Skills.
The package contains one file, SKILL.md. Once it is in place, the agent can use it in later sessions without further setup.
Install Biopython and set up NCBI access
The skill does not include the Biopython library. You need Python with Biopython installed before the agent's code will run. The source does not give an install command, so follow the official Biopython documentation for your environment.
Entrez access to NCBI requires an email address. The skill tells the agent to set Entrez.email before any NCBI request. An NCBI API key is optional and raises the rate limit from 3 to 10 requests per second, according to the guide.
from Bio import Entrez
Entrez.email = "your@email.com"Trigger the skill with natural requests
The skill is selected when your request matches its description and keywords, such as sequence, FASTA, genome, protein, BLAST, phylogenetic, biopython, bioinformatics, gene, DNA or RNA. Name the file type, the task and the goal in plain language. You do not need special syntax.
- Write a Biopython script that reads all records from genes.fasta and prints each GC content.
- Convert plasmid.gb to FASTA using Biopython.
- Translate this DNA sequence to protein and compute its molecular weight: ATGCGA...
- Search PubMed for 'CRISPR off-target' with Entrez and print the top 10 IDs.
- Read tree.nwk with Bio.Phylo, draw it, and give me the distance between two taxa.
- Parse 1abc.pdb and list the atoms in chain A.
Worked example: a typical session
Suppose you have a multi-record FASTA file and want the GC content of each sequence. A good request names the file, the operation and the output. The agent should then produce a script built on SeqIO.parse, which the skill describes as an iterator that is exhausted after one pass, and gc_fraction from Bio.SeqUtils. Review the code, then run it yourself in your own Python environment.
from Bio import SeqIO
from Bio.SeqUtils import gc_fraction
for rec in SeqIO.parse("file.fasta", "fasta"):
print(rec.id, round(gc_fraction(rec.seq), 3))Safety notes: scripts and network
This skill runs no scripts. It contains instructions and snippets only. The code the agent writes will run only if you run it. The skill does use the network: its BLAST and Entrez patterns send requests to NCBI services, and remote BLAST queries can take minutes. Check any script before you run it, and set a sensible timeout as the guide recommends.
Do not rely on results for clinical or medical decisions. The guide gives no validation or quality standards for such uses.
- Use SeqIO.index() instead of SeqIO.to_dict() for large files to avoid memory problems
- Add delays between batch NCBI requests to respect rate limits
- Filter BLAST hits by e-value (the guide suggests typically below 1e-5) and coverage
- Validate parsed data, since annotations are often missing in public databases
- Check that sequence alphabets match before running an operation
Tips for better results
Give the agent the file format and a sample of the file header or record when you can. Say which Biopython output you want, such as a new FASTA file or a printed table. If the code fails, paste the error message back, since the guide's pitfalls list covers common causes such as exhausted iterators and alphabet mismatches.
- Ask for one task per request, then combine scripts
- Tell the agent your email address for Entrez, or ask it to use a placeholder you fill in
- Ask the agent to explain each Biopython call so you can check it
What people use it for
- Compute GC content and reverse complements for a batch of FASTA sequences
- Convert GenBank records to FASTA for downstream tools
- Run an online BLAST search and filter hits by e-value
- Fetch nucleotide or PubMed records through Entrez with rate limits respected
- Read, draw and measure a Newick phylogenetic tree
- Parse a PDB or mmCIF file and compute distances between atoms
Frequently asked questions
Is the biology-biopython skill free to use?
The skill is published under the MIT licence, according to this listing. The skill itself runs no scripts and contains only a SKILL.md instruction file, so no separate fee is described in the source.
Does the skill install Biopython for me?
No. The skill does not bundle the Biopython library. You need Python with Biopython installed before the agent's code will run, and the skill does not give an install command.
Do I need an NCBI API key?
No. An email address is recommended for Entrez access, and the guide says to set it in code. An optional NCBI API key raises the rate limit from 3 to 10 requests per second.
Can the skill run BLAST searches for me?
It provides instructions and code for online BLAST through NCBIWWW and parsing with NCBIXML, but it does not run searches itself. Large BLAST jobs may take minutes, and you run the code yourself.
How is this different from just asking an AI for Biopython code?
The skill gives the agent a fixed set of standard Biopython patterns and common pitfalls, such as using SeqIO.index() for large files and handling SeqIO.parse as a one-pass iterator. Without it, the agent relies on general knowledge and may not follow these practices.
Why does my Biopython script stop after one loop?
SeqIO.parse returns an iterator that is exhausted after one pass, so a second loop over the same object finds nothing. Create the iterator again, or use SeqIO.index() for random access to large files.
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
Use this skill if you are a researcher, student or developer who wants an AI agent to write standard Biopython code for sequence analysis, file conversion, BLAST, NCBI lookups or tree handling. Pick something else if you need validated results for clinical decisions, or if you expect the skill to install Biopython, run analyses or handle large BLAST jobs for you.
Download the skill