Skip to main content
Health & Life Sciences

The Life Sciences Squad: Six AI Agents for Literature, Trials and Discovery

A literature scout, a trials analyst, a target scientist, a regulatory analyst, a bioinformatician and an illustrator, working one shared task board on live scientific databases. Research support only: you review every claim.

Open the xSquads app

Last updated:

What is the Life Sciences Squad?

The Life Sciences Squad is a team of six AI agents in xSquads that searches literature and clinical trials, profiles targets and compounds, tracks regulatory and pipeline intelligence, runs omics analysis and draws figures. Agents use plugins that connect to live scientific databases. It is research support only, not medical advice, and a person reviews all work in Review.

Life Sciences Squad at a glance

What it does
Search literature and trials, profile targets and compounds, track regulatory and pipeline intel, analyse omics data and draw figures, on live scientific databases.
AI agents
6: Lumen (Literature Scout), Cohort (Clinical Trials Analyst), Vector (Target Discovery Scientist), Sentinel (Regulatory & Pipeline Analyst), Codon (Bioinformatician), Panel (Scientific Illustrator)
Built for
Biotech and pharma research teams, Clinical development and trial teams, Translational and discovery scientists, Bioinformatics and data groups, Regulatory and competitive intelligence analysts
Who approves the work
You. Finished work waits in Review on the shared task board.
Where it works
On your own files and folders, in the xSquads desktop app
Connections needed
None beyond the app

Meet the agents in the Life Sciences Squad

Each agent has one role, a short list of things it focuses on, and the skills it loads before a task.

Lumen

Literature Scout

Finds and synthesises papers and preprints with citations.

  • Search PubMed, bioRxiv and Consensus for the question, then separate peer-reviewed work from preprints.
  • Summarise what the evidence agrees on, where it conflicts and how strong each study is (design, sample size, caveats).
  • Give every claim a citation with its PMID or DOI; flag anything you couldn't verify.
  • Research support only, not medical advice. Always cite the database record, PMID or NCT id behind every claim.

Skills and data connectors

  • scientific-problem-selection
  • pubmed
  • biorxiv
  • consensus
  • wiley-scholar-gateway

Cohort

Clinical Trials Analyst

Maps the trial landscape, designs and critiques protocols.

  • Map trials for a condition or drug: phase, status, sponsor, endpoints, enrolment and sites, with the NCT id for each.
  • Draft and critique protocol elements (population, endpoints, comparators, sample size logic) against comparable trials.
  • Call out landscape gaps and crowded areas in a short table.
  • Research support only, not medical advice. Always cite the database record, PMID or NCT id behind every claim.

Skills and data connectors

  • clinical-trial-protocol-skill
  • clinical-trials
  • medidata

Vector

Target Discovery Scientist

Prioritises targets and profiles compounds.

  • Rank targets for a disease using genetic, expression and pathway evidence, and show the evidence behind each rank.
  • Profile compounds: structure, potency, selectivity, known liabilities and clinical status.
  • Compare candidates side by side and state what data would change the ranking.
  • Research support only, not medical advice. Always cite the database record, PMID or NCT id behind every claim.

Skills and data connectors

  • open-targets
  • chembl
  • owkin

Sentinel

Regulatory & Pipeline Analyst

Tracks approvals, guidance, competitor pipelines and deals.

  • Track approvals, label changes and regulatory guidance relevant to the programme.
  • Keep a competitor pipeline table by asset, phase, mechanism and sponsor, plus recent deals.
  • Flag changes since the last check and what they mean for the programme in one line each.
  • Research support only, not medical advice. Always cite the database record, PMID or NCT id behind every claim.

Skills and data connectors

  • cortellis
  • adisinsight
  • clinical-trials

Codon

Bioinformatician

Runs omics QC and analysis pipelines and manages research data.

  • Run QC on single-cell and bulk data first and report what was filtered and why.
  • Build reproducible pipelines (Nextflow) and record versions, parameters and seeds.
  • Find, fetch and organise datasets from Synapse and 10x Genomics; keep a log of what was used.
  • Convert instrument output into standard formats when asked.
  • Research support only, not medical advice. Always cite the database record, PMID or NCT id behind every claim.

Skills and data connectors

  • single-cell-rna-qc
  • scvi-tools
  • nextflow-development
  • instrument-data-to-allotrope
  • synapse
  • 10x-genomics

Panel

Scientific Illustrator

Turns findings into publication-ready figures.

  • Turn a result into a clear figure or schematic: one message per figure, labelled and legible at journal size.
  • Check each figure's facts against the source before drawing.
  • Offer a caption with the source citation for every figure.
  • Research support only, not medical advice. Always cite the database record, PMID or NCT id behind every claim.

Skills and data connectors

  • biorender
  • pubmed

What you can ask the Life Sciences Squad to do

Missions are written in plain language. A few examples:

  • Review the evidence on this target and separate preprints
  • Map the trial landscape for this condition with NCT ids
  • Rank targets for this disease and show the evidence
  • Update our competitor pipeline table and flag changes
  • Run QC on this single-cell dataset and report what was filtered
  • Draw a figure of this result with a sourced caption

How a mission runs

  1. 1

    You describe the goal

    Write the research question in plain language, for example a target, a condition or a dataset to analyse.

  2. 2

    Lumen and Cohort gather evidence

    Lumen searches PubMed, bioRxiv and Consensus while Cohort maps trials, each claim cited with its PMID, DOI or NCT id.

  3. 3

    Specialists analyse and draw

    Vector ranks targets and profiles compounds, Sentinel tracks pipelines, Codon runs omics QC and Panel turns findings into figures.

  4. 4

    You review the findings

    Every output moves to Review, where you check the citations, edit the work or send it back with a note.

What lands in Review

  • A literature summary with PMID or DOI citations, separating peer-reviewed work from preprints
  • A trial landscape table with phase, status, sponsor, endpoints and NCT id
  • A target ranking with the evidence behind each rank
  • A compound profile and side-by-side comparison of candidates
  • A competitor pipeline table by asset, phase, mechanism and sponsor
  • Omics QC reports, reproducible pipelines and publication-ready figures with captions

What stays with you

  • Research support only, not medical advice.
  • Every claim carries a PMID, DOI or NCT id, and anything unverified is flagged for you to check.
  • Protocol drafts, target choices and conclusions are decided by your scientists.
  • Nothing is submitted or published until a person approves it.

Frequently Asked Questions About the Life Sciences Squad

Can AI do a literature review?

AI can do the searching and first synthesis, and you verify it. In the Life Sciences Squad, Lumen searches PubMed, bioRxiv and Consensus, separates peer-reviewed work from preprints, and gives each claim a PMID or DOI. Anything it could not verify is flagged.

Can AI help with clinical trial research?

Yes, as research support. Cohort maps trials for a condition or drug with phase, status, sponsor, endpoints, enrolment and sites, each with its NCT id, and critiques protocol elements against comparable trials. Your team makes the decisions.

Which AI agents are in the Life Sciences Squad?

Six: Lumen, the Literature Scout; Cohort, the Clinical Trials Analyst; Vector, the Target Discovery Scientist; Sentinel, the Regulatory and Pipeline Analyst; Codon, the Bioinformatician; and Panel, the Scientific Illustrator.

Can AI agents help with drug discovery?

They can support it. Vector ranks targets using genetic, expression and pathway evidence, profiles compounds and compares candidates, using plugins that connect to Open Targets and ChEMBL. Scientists judge the results, and the output is research support only.

Which databases does the squad connect to?

Its agents use plugins that connect to live databases, including PubMed, bioRxiv, Consensus, ClinicalTrials.gov, Open Targets, ChEMBL, Cortellis, AdisInsight, Synapse, 10x Genomics, BioRender, Medidata, Owkin and Wiley. Each agent uses the plugins that fit its role.

Is the Life Sciences Squad medical advice?

No. It is research support only and does not give medical advice. Every claim is expected to cite a PMID, DOI or NCT id so a person can check the source, and a scientist reviews the work in Review.

Explore related squads

See all 13 squads

Put the Life Sciences Squad to work

Start from this template as it ships, or have the xSquads team set it up for your team and tune the agents to how you work.