Epiphytic

Run nf-core bioinformatics pipelines (rnaseq, sarek, atacseq) on sequencing data. Use when analyzing RNA-seq, WGS/WES, or ATAC-seq data—either local FASTQs or public datasets from GEO/SRA. Triggers on nf-core, Nextflow, FASTQ analysis, variant calling, gene expression, differential expression, GEO reanalysis, GSE/GSM/SRR accessions, or samplesheet creation.

Simple task management using a shared TASKS.md file. Reference this when the user asks about their tasks, wants to add/complete tasks, or needs help tracking commitments.

Use when you need direct browser control - teaches Chrome DevTools Protocol for controlling existing browser sessions, multi-tab management, form automation, and content extraction via use_browser MCP tool

Use when working on Claude Code plugins (creating, modifying, testing, releasing, or maintaining) - provides streamlined workflows, patterns, and examples for the complete plugin lifecycle

Use when working with Claude Code CLI, plugins, hooks, MCP servers, skills, configuration, or any Claude Code feature - provides comprehensive official documentation for all aspects of Claude Code

Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies

Use when starting any conversation - establishes how to find and use skills, requiring Skill tool invocation before ANY response including clarifying questions

Use when creating new skills, editing existing skills, or verifying skills work before deployment

This skill should be used when the user asks to "run the brains pipeline", "start the brains workflow", "plan and implement from scratch", "do an ADR", "start with brainstorming", or invokes "/brains:brains". Phase 1 of the BRAINS pipeline: interactive research, question generation, questionnaire, architecture synthesis, and ADR production. Supports --single, --parallel (default), and --debate modes, and an optional --autopilot flag that auto-chains into hands-off map + implement. Chains into /brains:map at the user gate.

This skill should be used when the user asks to "document mode", "edit only docs", "just update the docs", "fast-path a doc change", or invokes "/brains:document". Canonical entry point for BRAINS document mode: an abbreviated spine for document-only changes that skips planning, teammate orchestration, nurture, and secure. Runs an eligibility gate, lightweight research, the full 2-4 question questionnaire, a slim ADR, inline document edits, direct star-chamber council review (replacing nurture+secure), and an inline commit. Supports --single, --parallel (default), and --debate modes for the review phase, plus --autopilot and --lean.

This skill should be activated when a task involves significant complexity — multiple components, cross-cutting concerns, unfamiliar codebases, architectural decisions, or multi-step implementations. It suggests the BRAINS workflow to the user without auto-invoking it.

This skill should be used when the user asks to "set up brains", "configure brains", "install brains dependencies", "set up multi-LLM", "configure star-chamber for brains", "change brains defaults", "set up kroki", "enable diagram rendering", or invokes "/brains:setup". Guides the user through installing dependencies, configuring LLM providers, setting default modes for each BRAINS skill, and optionally starting a local Kroki container for diagram rendering.

This skill should be used when the user asks to "nurture", "review and fix", "improve code quality", "add tests", "fix remaining bugs", "polish the implementation", "check for completeness", or invokes "/brains:nurture". Reviews implementation for completeness, then fixes issues including bugs, missing features, and lack of tests. Supports --single (default), --parallel, and --debate modes for the review phase. When invoked by a phase-3 teammate with --scope phase-N, scopes the review to changes in that plan-phase, commits code and updates .gitignore, reflects any half-complete state in docs, and files follow-up beads tasks for the next phase or cleanup.

This skill should be used when the user asks to "implement the plan", "execute the plan", "start implementation", "build the tasks", "run the teammates", or invokes "/brains:implement". Phase 3 of the BRAINS pipeline: spawns a teammate Claude Code instance per plan-phase via agent-teams (preferred) or tmux (fallback), waits on beads state, handles task failures with two-strike-plus-human-in-loop flow. Supports --single, --parallel (default), and --debate modes for nurture/secure review within each plan-phase, plus an optional --autopilot flag that runs hands-off across phases until a needs-human ticket or direct user intervention stops it. Also supports --resume to pick up after a pause.

This skill should be used when the user asks to "map out the plan", "create the implementation plan", "outline the tasks", "plan it out", "sketch the implementation", or invokes "/brains:map". Phase 2 of the BRAINS pipeline: high-level plan generation (no implementation specifics), topic-slug derivation, optional branch creation, beads task creation with brains:-prefixed labels, and user approval gate. Supports --single, --parallel (default), and --debate modes for plan review, plus an optional --autopilot flag that skips user gates and auto-chains into /brains:implement --autopilot. Chains into /brains:implement at the user gate.

Generate architecture diagrams for ADRs. Emits Mermaid source (.mmd) for flowchart, state, er, and sequence types and Structurizr DSL (.dsl) for c4. Source and .svg output are stored side-by-side in docs/adr/diagrams/. Invokable standalone or auto-triggered from brains:brains phase-1 step 8.

Autonomous development orchestrator that plans, builds, and validates applications

Spawn a sandboxed LLM with multi-LLM coordination for automated code review

Spawn a sandboxed LLM instance to work on a delegated task

Multi-agent git workflow orchestrator with isolated worktrees, FIFO merge queue, and automated PR lifecycle management.