Skill catalog¶
Each skill is a folder with a SKILL.md that an agent loads when a task matches its description. This page is generated from each skill's frontmatter when the site is built.
| Skill | What it does | Version | Evaluation |
|---|---|---|---|
| bulk-rnaseq | Find top differentially expressed genes from bulk RNA-seq counts | 0.6 | Report |
| open-weight-models | Package, register, validate & deploy open-weight HLS models on Databricks — e.g | 0.5 | Pending |
| pathway-enrichment-analysis | Pathway enrichment for RNA-seq (ORA or GSEA) | 0.5 | Pending |
| sc-rnaseq | Analyze single-cell RNA-seq data from .h5ad files | 0.5 | Evidence |
| cohort-builder | Build a reproducible, feasibility-checked patient cohort from coded + free-text clinical data | 0.5 | Report |
| phi-deidentifier | HIPAA Safe-Harbor de-identify a Unity Catalog table (k-anonymity + governed view) | 0.5 | Report |
| rwe-cohortstudy | Comparative effectiveness (cohort study) with propensity adjustment (matching, IPW) | 0.5 | Report |
| semantic-builder | Canonical payer+provider measure catalog mapped to Unity Catalog metric views | 0.5 | Pending |
| accelerators | Route imaging and biology-model tasks to an HLS accelerator — Pixels or Genesis Workbench — with the exact next step | 0.5 | Report |
| presentation-builder | Turn a cohort, dataset, model, or analysis into a grounded, shareable HTML deck (PHI de-id delegated to phi-deidentifier) | 0.1 | Pending |
| report-builder | LaTeX report generation and PDF compilation from analysis notebooks via Tectonic | 0.5 | Pending |
| scientific-figures | Publication-quality scientific and biopharma figures for notebooks, apps, and reports | 0.5 | Pending |
| skill-eval | Benchmark a skill: paired runs with/without it, MLflow genai.evaluate scoring, win/loss + failure taxonomy | 0.5 | Pending |
Evaluation: Report links a paired with/without-skill comparison, Evidence links benchmark files without a report yet, and Pending means no evaluation has been committed. See Evaluation.