BioRouter AI Agent Marketplace
One-click downloads for BioRouter extensions, community workflows, and reusable skills.
BioRouter extensions
.brxt · One-click installDownload a .brxt bundle and drag it into BioRouter, Extensions, Add Extension. BioRouter installs the virtual environment, configures the tool, and adds it to your session. No terminal needed.
Multimodal access to the UCSF Clinical Data Warehouse through natural language. One-call cohort building across diagnoses, medications, procedures, labs, radiology/imaging, immunizations, allergies, and vitals; clinical-notes/NLP search; read-only queries, schema discovery, and structured results. Requires UCSF network credentials (CAMPUS\username and password).
Natural language SQL on the UCSF OMOP de-identified clinical database. Read-only access to standardized EHR data under the OMOP Common Data Model. v0.2.0 adds built-in OMOP/SQL-Server context plus concept-search, lab-finder, and live schema tools so the agent reaches the right data faster and more reliably. Requires UCSF credentials.
Browser automation via Microsoft's @playwright/mcp. Navigate websites, extract data, fill forms, and run web research workflows using structured accessibility snapshots. No vision model required. Requires Node.js on the host.
Structure-aware access to the SPOKE biomedical knowledge graph (43M nodes): live schema introspection, entity/identifier resolution, node profiling, shortest-path finding, and guarded read-only Cypher across diseases, genes, proteins, drugs, and pathways. Includes a bundled spoke-knowledge-graph skill. Requires a SPOKEAGENT_PASSCODE (see credentials page).
Pre-indexed code knowledge graph. Ask "who calls X?", "what does Y call?", or "what breaks if I change Z?" across 23 languages including R, Julia, MATLAB, Perl. Vendored fork of CodeGraph on tree-sitter.
Create, read, and edit Word (.docx), Excel (.xlsx), and PowerPoint (.pptx) files from chat. Handles headings, tables, charts, and 150+ Excel functions with live formula evaluation. Built on the open-source OfficeCLI engine, so no Microsoft Office install is needed. Ships four bundled skills for the office suite, Word, Excel, and PowerPoint.
Community workflows
All on GitHub →Query an OMOP-based EHR to identify a Type 2 Diabetes cohort, compute demographic summaries (gender, age, race, ethnicity), and auto-generate a shareable HTML report and interactive R Shiny dashboard. Runs at any OMOP-compatible institution without moving patient data.
version: 1.0.0 title: EHR Diabetes Demographics and Reporting Dashboard description: A workflow for querying an OMOP-based EHR to derive a disease cohort (e.g., T2D), summarize demographics, and produce sharable visual outputs. instructions: | Use an OMOP CDM clinical database to identify a patient cohort via diagnosis concepts. Join to person to obtain demographics and compute counts/percentages. Present results as an HTML report and optional R Shiny dashboard. activities: - Count disease cohort - Demographic breakdowns - Generate HTML report - Build Shiny dashboard extensions: []
.zip below, then in BioRouter open Skills → Add Skill and drag the .zip (or a single SKILL.md) into the drop zone. BioRouter unpacks the bundle into ~/.config/biorouter/skills/ and registers every SKILL.md inside. You can also unzip manually and copy the folder into ~/.config/biorouter/skills/. Skills appear under BioRouter → Skills automatically; auto-applied skills activate when relevant files are touched.
End-to-end research: scopes the question, reviews real literature, generates hypotheses, plans and verifies the analysis, then writes a cited draft. Delegates biomedical data to BioSkills and the SPOKE/CDW/OMOP extensions. It never fabricates a citation or result.
Anti-slop frontend design for landing pages, dashboards, lab tools, and paper companion sites. Reads the brief, picks a real design direction, and ships interfaces that don't look templated.
A pass/fail checklist for spotting AI writing patterns before generating prose, articles, or essays.
Applies ggplot2 best-practice style when writing R plotting code.
Applies tidyverse conventions and documentation standards when writing or reviewing R code.
Applies Python naming, typing, error handling, and project structure conventions when writing Python code.
End-to-end planning for the Ralph autonomous agent loop. Drafts a markdown PRD from a feature idea, then converts it to prd.json. Each story is sized for one hands-off iteration.
Engineering discipline skills: brainstorming, TDD, systematic debugging, parallel agents, plan writing, code review, git worktrees. Workflows tested on long-running engineering tasks.
SSH setup, SLURM job templates (CPU/GPU/H200), file transfer, module management, and common pitfalls for the UCSF CHPC cluster.
Multi-stage, confidence-scored review of a pull request or local diff. Drops low-confidence findings and posts a formatted PR comment via gh.
Simplifies recently-changed code for clarity and consistency while preserving exact behavior. Defers language style to the Python and R skills.
Three git workflows: a clean single commit; commit + push + open a PR; and cleanup of [gone] branches and their worktrees.
The canonical guide to authoring BioRouter skills: frontmatter, progressive disclosure, packaging, and testing.
Turns project guardrails (warn/block rules) into real BioRouter hooks in .biorouter/hooks.yaml. A block rule stops the action; a warn rule shows a notice. Requires allow_project_hooks: true.
Phased legacy-modernization workflow (preflight, assess, map, extract rules, brief, transform, harden, status) for COBOL, legacy Java/C++, and monoliths.
Generates self-contained single-file interactive HTML playgrounds for configuring something visually, then copying out a ready-to-use prompt.
Anthropic's anti-slop frontend guidance for distinctive, production-grade UIs. Complements the Taste skill above.
Multiple sequence alignment and pairwise alignment.
SAM/BAM/CRAM manipulation, sorting, indexing, dedup, and stats.
RNA isoform and splice-junction analysis from short- and long-read data.
Chromatin accessibility, peak calling, footprinting, and deep-learning ATAC.
Mendelian randomization, colocalization, fine-mapping, and TWAS.
Molecular structure handling, descriptors, and drug-discovery utilities.
Transcription-factor and histone-mark peak calling and differential binding.
Survival, mixed models, and clinical-trial statistical analysis.
ClinVar, OMIM, COSMIC, gnomAD, and phenotype DB queries.
Protein–RNA interaction mapping from iCLIP/eCLIP/PAR-CLIP data.
Cross-species synteny, orthology, and selection scans.
CNV detection, visualization, and segmentation.
Pooled screen counting, MAGeCK analysis, and hit prioritization.
Publication-quality plots: heatmaps, volcano, Manhattan, dimplots.
NCBI, UniProt, Ensembl, and biological database APIs.
RNA-seq DE testing with DESeq2, edgeR, and limma-voom.
Environmental DNA, ecological population genomics, and biodiversity.
Pathogen typing, outbreak phylogeography, and surveillance.
m6A and other RNA modification detection from sequencing.
Power analysis and sample-size calculation for omics studies.
Count normalization, gene ID mapping, and matrix QC.
FCS parsing, gating, and high-dimensional cytometry analysis.
TF–target network inference from expression and chromatin data.
Gene prediction, repeat masking, and functional annotation.
De novo assembly, polishing, and assembly QC.
CRISPR guide design, off-target prediction, and editing analysis.
BED/GTF arithmetic, overlap, and feature manipulation.
3D chromatin contacts, TAD calling, and loop detection.
Single-cell spatial proteomics from IMC and MIBI data.
MHC binding, neoantigen prediction, and HLA typing.
Cell-free DNA fragmentomics, tumor fraction, and methylation cfDNA.
ONT and PacBio basecalling, alignment, and structural-variant workflows.
Biomarker discovery, model training, and cross-validation pipelines.
Mass-spec metabolite identification, alignment, and quantification.
Shotgun microbial profiling, assembly, and binning.
Bisulfite alignment and differential methylation testing.
16S rRNA amplicon and microbiota community profiling.
Cross-modality fusion: MOFA, DIABLO, WNN, totalVI.
GO, KEGG, Reactome, and GSEA enrichment testing.
Haplotype phasing and genotype imputation with SHAPEIT and Beagle.
Tree construction with IQ-TREE, RAxML, BEAST, and ancestral state.
GWAS, population structure, LD, and selection scans.
PCR and qPCR primer generation with Primer3 and primerBLAST.
Mass-spec quantification and protein abundance pipelines.
Short-read mapping with BWA-MEM2, Bowtie2, STAR, HISAT2.
FastQC, fastp, trimming, and contamination screening.
Reproducible HTML, Quarto, and R Markdown reports.
Restriction-enzyme mapping and digestion prediction.
Ribosome profiling and translation-efficiency analysis.
Gene and transcript abundance with Salmon, kallisto, RSEM.
Secondary-structure prediction with RNAfold, IPknot.
FASTA/FASTQ parsing, conversion, and indexing.
Transcription, translation, ORF finding, and motif search.
scRNA-seq clustering, annotation, trajectory, and integration.
miRNA, piRNA, and small-RNA quantification.
Visium, Slide-seq, and high-resolution tissue expression.
PDB parsing, AlphaFold, ESMFold, and structure analysis.
Metabolic flux balance, kinetic modelling, network analysis.
Immune-receptor repertoire profiling and clonal-family analysis.
Time-series and circadian expression analysis.
SNPs, indels, SVs with GATK, DeepVariant, Manta, Delly.
Snakemake, Nextflow, CWL, WDL pipeline scaffolding.
Ready-to-run analysis pipelines combining multiple skills.
Have an agent, workflow, or skill to share with the BioRouter community?
Contribute on GitHub