Query 40+ bioinformatics databases (UniProt, KEGG, ChEMBL, Reactome) through one consistent Python interface.
An end-to-end router that takes bulk RNA-seq from raw FASTQ through QC, quantification, a counts matrix, DESeq2, pathway enrichment, and publication figures.
Guides Claude to write correct Astropy code for units, coordinate transforms, FITS I/O, tables, time scales, WCS, and cosmology.
Runs a long-horizon experiment loop that improves a code/model/prompt artifact against an evaluator, using a persistent hypothesis tree and a held-out test merge gate to avoid overfitting.
Decides which regulatory framework governs (ICH Q2(R2)/Q14, USP, ICH M10, CLSI EP, ISO 17025), then designs the study, computes the statistics correctly, and structures the validation, verification, or transfer documentation.
A skill for organizing, querying, validating and converting neuroscience data to the BIDS standard.
Teaches Claude to use the aeon 1.x toolkit correctly for time series classification, regression, clustering, forecasting and anomaly detection.
Clusters a GitHub issue backlog by root cause into a handful of plan-master issues plus design docs, then bundles PRs that close each cluster atomically.
A hands-on playbook for validating web accessibility with VoiceOver, NVDA, JAWS and TalkBack.
Operate Gemini Notebook (NotebookLM) end to end via the notebooklm-py CLI or typed async Python API — auth, sources, grounded chat, research, and artifact generation.
Auto-detects any knowledge source — docs site, GitHub repo, PDF, video, codebase — and turns it into a packaged AI skill.
A router skill that dispatches any UI/UX request to four rulesets — taste, brand design systems, performance, and design ops — so output looks like something a senior Linear/Stripe designer would ship.
Turns Claude into an explore-first planning consultant that grounds in your codebase, asks only the decisions you must own, and — only after your explicit okay — writes one decision-complete work plan.
A post-implementation review gate that spawns five parallel sub-agents — goal verification, hands-on QA, code quality, security, and context mining — and passes only if all five pass.
A four-language engineering constitution that forces reference reading before coding, then enforces strict types, TDD, and file-size discipline.
A skill that analyzes CSV, Parquet, and JSON data by intelligently routing each operation to either DuckDB or Polars for maximum speed.
Scans a repository and auto-generates a hierarchical set of AGENTS.md files at the root and in complexity-scored subdirectories.
Drives the agent-browser CLI so Claude can navigate sites, fill forms, capture screenshots and extract page data.
Search and refactor code by syntax structure (AST) instead of text across 25 languages, with dry-run-first codemods.
Turns a single document into a kawaii-styled NotebookLM slide-deck PDF plus a mobile-friendly share link.