Arjun Raj's systems biology lab
Official@arjunrajlaboratory
Arjun Raj's systems biology lab maintains Mycelium, a living-repository plugin for reproducible bioinformatics analysis, data provenance, and scientific reporting.
Agent Skills by Arjun Raj's systems biology lab
Showing 10 vetted skills indexed across 1 GitHub repositories.
arjunrajlaboratorydevelop
Develop and verify the Mycelium plugin across Claude Code and Codex hosts.
arjunrajlaboratorycore
Initialize and maintain Mycelium living repositories with structured project memory.
arjunrajlaboratoryingest
Import datasets into a Mycelium project with metadata, provenance, and manifest registration.
arjunrajlaboratorylifecycle-audit
Runs black-box lifecycle smoke tests of Mycelium hooks through real Claude Code or Codex CLI tasks.
arjunrajlaboratorytransfer
Transfers actionable learnings between sibling projects' .living/learnings.md files.
arjunrajlaboratoryreview
Review analytical code and scientific changes using six specialized checklists and synthesized findings.
arjunrajlaboratoryanalyze
Run convention-aware computational analyses with registered datasets, linting, and reproducible pipelines.
arjunrajlaboratoryreport
Generate structured reports and PDFs from completed Mycelium analyses using installed convention packs.
arjunrajlaboratoryideas
Generates research ideas by routing brainstorming sessions through persona-based convention packs.
arjunrajlaboratorycodex-review
Address Codex pull request comments and audit the branch for repeated error patterns.
Frequently Asked Questions About Arjun Raj's systems biology lab
FAQPage SchemaWhat tasks can I perform with the Raj lab's Mycelium skills?▼
You can initialize living repositories, ingest datasets (CSV, FASTQ, XLSX, TIFF, FCS) with provenance tracking, run statistical and bioinformatics analyses like differential expression and UMAP, review analytical code, cross-pollinate learnings between subprojects, brainstorm research ideas, and produce shareable PDF reports.
Who are the Mycelium skills designed for?▼
They target computational biologists, bioinformaticians, and data scientists working in Claude Code or Codex environments who need reproducible analysis pipelines, dataset provenance, statistical review checklists, and publication-ready reporting within a versioned living-repository structure.
How does Mycelium handle data ingestion and provenance?▼
The ingest skill adds files and datasets with metadata, validation, and DATA_MANIFEST updates, supporting collaborator or facility outputs in CSV, FASTQ, XLSX, TIFF, JSON, FCS, and TSV formats. It deliberately excludes analysis, deletion, or URL downloads to keep provenance records clean.
What review and quality-control capabilities are included?▼
The review skill audits analytical code and scientific changes using bundled statistical, data-leakage, bioinformatics, documentation, and code-quality checklists, producing prioritized findings. The codex-review skill separately generalizes Codex PR comments into error patterns and audits the whole branch for recurrences.
How is the Mycelium plugin itself tested and maintained?▼
The develop skill covers plugin manifests, lifecycle hooks, installers, and cross-host compatibility with TDD and smoke testing. The lifecycle-audit skill runs black-box lifecycle tests through real Claude Code or Codex sessions, verifying hook firing, state transitions, and scientific-tree isolation before merging.