# Omics-OS Documentation

- [Introduction](/docs): Documentation for Omics-OS — The Operating System for Biology. Workspace, OmniBar, Playground, Editor, Cloud CLI, and Lobster AI Engine. BioVault is coming soon.
- [Agent Coordination Quick Reference](/docs/advanced/agent-coordination-ref): - /Users/tyo/GITHUB/omics-os/lobster/lobster/agents/graph.py - Delegation tool creation - /Users/tyo/GITHUB/omics-os/lobster/lobster/agents/supervisor.py...
- [Agent Coordination Patterns for Metadata Operations](/docs/advanced/agent-coordination): This document describes how agents coordinate around metadata operations in the Lobster system, including handoff patterns, permission flows, and data sharin...
- [Custom Feature Agent - AI-Powered Feature Generation](/docs/advanced/custom-feature-agent): The Custom Feature Agent is a META-AGENT that leverages the Claude Code SDK to automatically generate production-ready Lobster components through natural...
- [Data Expert Agent Enhancements v0.2+](/docs/advanced/data-expert-enhancements): The Data Expert Agent has been significantly enhanced to provide comprehensive workspace management capabilities alongside its core data acquisition and proc...
- [Docker Deployment Guide (Experimental)](/docs/advanced/docker-deployment): Comprehensive guide to running Lobster AI in Docker containers for development, staging, and production environments.
- [Download Queue System](/docs/advanced/download-queue): The Download Queue System is a structured handoff contract between researchagent and dataexpert, enabling coordinated dataset downloads with pre-valida...
- [Agent-Guided Formula Construction Documentation](/docs/advanced/formula-construction): The agent-guided formula construction feature enhances the transcriptomicsexpert agent with intelligent statistical guidance for differential expression a...
- [GEO Download System Improvements (November 2024)](/docs/advanced/geo-improvements): This document details critical improvements to the GEO download system implemented in November 2024 to address three major bugs identified by Kevin's testing...
- [Manual Cell Type Annotation Service Documentation](/docs/advanced/manual-annotation): The Manual Cell Type Annotation Service provides expert-guided cell type annotation capabilities for single-cell RNA-seq data with a color-synchronized Rich ...
- [Metadata Update Workflows](/docs/advanced/metadata-workflows): This document traces the complete metadata lifecycle from creation through enrichment to export, documenting how metadata flows through Lobster's multi-agent...
- [Microbiome Metadata Harmonization Workflow](/docs/advanced/microbiome): The Microbiome Metadata Harmonization Workflow enables automated harmonization of microbiome metadata from publication references to filtered, analysis-r...
- [Migration Guide (v1.0.0)](/docs/advanced/migration-v0-to-v1): Migrating custom packages to Lobster AI v1.0.0 modular architecture
- [Multi-Omics Integration](/docs/advanced/multi-omics): Multi-omics integration combines data from multiple biological measurement modalities (transcriptomics, proteomics, metabolomics, etc.) to gain comprehensive...
- [Protein Structure Visualization Expert Agent](/docs/advanced/protein-structure): Since v0.2 - Protein structure analysis with PyMOL visualization and BioPython integration
- [Publication Content Access & Provider Architecture](/docs/advanced/publication-intelligence): Version: 2.4.0+ (Phase 1-6 Refactoring Complete) Status: Production-ready Implementation: ContentAccessService with Provider Infrastructure (Janu...
- [Supervisor Configuration Guide](/docs/advanced/supervisor-config): The Lobster supervisor agent now features a dynamic, configuration-driven prompt system that automatically discovers agents and their capabilities, eliminati...
- [Vector Search Internals](/docs/advanced/vector-search-internals): Deep dive into the vector search architecture: backends, embedders, rerankers, ontology collections, and S3 auto-download
- [Workspace Content Service](/docs/advanced/workspace-content): The WorkspaceContentService provides structured, type-safe caching of research content (publications, datasets, metadata) in the DataManagerV2 workspace....
- [Drug Discovery](/docs/agents/drug-discovery): Drug target identification, compound profiling, clinical translation, and pharmacogenomics agent
- [Genomics](/docs/agents/genomics): Variant analysis and GWAS agent
- [Agent Packages](/docs/agents): Domain-specific agent packages for Lobster AI
- [Metabolomics](/docs/agents/metabolomics): LC-MS, GC-MS, and NMR data analysis — QC, preprocessing, multivariate statistics, metabolite annotation, and pathway enrichment
- [Metadata](/docs/agents/metadata): Publication queue processing and metadata filtering agent
- [Machine Learning](/docs/agents/ml): ML data preparation, feature selection, survival analysis, and interpretability
- [Proteomics](/docs/agents/proteomics): Mass spectrometry and protein analysis agent
- [Research](/docs/agents/research): Data discovery and research planning agents
- [Structural Visualization](/docs/agents/structural-viz): Protein structure visualization and analysis agent
- [Transcriptomics](/docs/agents/transcriptomics): Single-cell and bulk RNA-seq analysis agents
- [Visualization](/docs/agents/visualization): Single-cell and omics visualization agent
- [Agents API Reference](/docs/api-reference/agents): The Agents API provides specialized AI agents for different analytical domains in bioinformatics. Each agent is designed as an expert in its specific area, o...
- [Core API Reference](/docs/api-reference/core): The Core API provides the foundational layer of the Lobster AI system, including data management, client interfaces, and system orchestration. This module ha...
- [Interfaces API Reference](/docs/api-reference/interfaces): The Interfaces API defines the abstract contracts and protocols that ensure consistent behavior across different implementations in the Lobster AI system. Th...
- [API Reference Overview](/docs/api-reference/overview): The Lobster AI API provides a comprehensive set of interfaces for multi-omics bioinformatics analysis through a professional agent-based architecture. This r...
- [Services API Reference](/docs/api-reference/services): The Services API provides stateless analysis services implementing scientific algorithms for bioinformatics workflows. All services follow the stateless patt...
- [Architecture Overview](/docs/architecture/overview): Lobster AI is a modular bioinformatics platform with pluggable execution environments, LLM providers, and integrated data management. The platform archit...
- [Drug Resistance: Overcoming the BCR-ABL T315I Gatekeeper in CML](/docs/case-studies/cml-resistance): From target validation to resistance pharmacogenomics — how Lobster AI analyzes the canonical drug resistance story in chronic myeloid leukemia.
- [Drug Discovery: ALK Inhibitor Investment Analysis](/docs/case-studies/drug-discovery): Multi-agent drug target validation, compound profiling, and resistance pharmacogenomics using Lobster AI's drug discovery agents.
- [Genomics: From Variant QC to Clinical Prioritization](/docs/case-studies/genomics): Population and clinical genomics across three complexity levels — variant QC, mitochondrial annotation, and a full clinical prioritization pipeline.
- [Case Studies](/docs/case-studies): Real-world analysis workflows with Lobster AI
- [Machine Learning: From Data Preparation to Survival Prediction](/docs/case-studies/machine-learning): ML pipelines across three complexity levels — data preparation, biomarker discovery with bootstrap feature importance, and Cox proportional hazards survival analysis.
- [Metabolomics: From LC-MS Quality Control to NMR Pathway Profiling](/docs/case-studies/metabolomics): Metabolomics analysis across three complexity levels — LC-MS plasma QC, NMR dataset discovery, and type 2 diabetes metabolic profiling with pathway annotation.
- [Proteomics: From DIA-MS Quality Control to Biomarker Discovery](/docs/case-studies/proteomics): Mass spectrometry proteomics across three complexity levels — CSF quality control, tumor differential expression, and clinical biomarker panel discovery.
- [Research: From Literature Mining to Dataset Discovery](/docs/case-studies/research): Literature search and dataset discovery across three complexity levels — CRISPR base editing review, spatial transcriptomics datasets, and single-cell multi-omics method comparison.
- [Transcriptomics: From Single-Cell QC to Pseudobulk Differential Expression](/docs/case-studies/transcriptomics): Single-cell and bulk RNA-seq analysis across three complexity levels — QC, cell type annotation, and pseudobulk DE with pathway enrichment.
- [BioVault](/docs/cloud/biovault): Evidence synthesis from biological sources
- [Lobster CLI](/docs/cloud/cli): Connect to Omics-OS Cloud from your terminal
- [Cockpit](/docs/cloud/cockpit): Your launch pad into Omics-OS Cloud
- [Omics-OS Cloud](/docs/cloud): Cloud platform for biological data analysis — Workspace, Playground, Editor, BioVault, and CLI
- [Workspace](/docs/cloud/workspace): Canvas-based analysis environment with AI assistant
- [Component Registry](/docs/core/component-registry): Plugin discovery via entry points with ComponentRegistry
- [Data Manager](/docs/core/data-manager): Modality storage and workspace management with DataManagerV2
- [Core SDK](/docs/core): The lobster-ai core package - foundation for all agent packages
- [Service Protocols](/docs/core/protocols): Duck-typed interfaces for services with ServiceProtocol
- [Provenance Tracking](/docs/core/provenance): W3C-PROV compliant analysis tracking with ProvenanceTracker
- [Advanced Agent Customization](/docs/developer/agent-customization): This guide covers advanced agent customization in Lobster AI, enabling you to build production-ready specialized agents that seamlessly integrate with th...
- [Creating Adapters - Lobster AI Adapter Development Guide](/docs/developer/creating-adapters): This guide covers how to create adapters in the Lobster AI system. Adapters serve two primary purposes: Modality Adapters convert raw data from various s...
- [Creating Agent Packages](/docs/developer/creating-agents): Comprehensive guide to building Lobster AI agent packages with step-by-step instructions and real examples
- [Creating Services - Lobster AI Service Development Guide](/docs/developer/creating-services): This guide covers how to create stateless analysis services in the Lobster AI system. Services handle the core computational work for bioinformatics analyses...
- [Custom Provider Development Guide](/docs/developer/custom-providers): Custom Providers enable Lobster AI to integrate with external data sources, APIs, and databases beyond the built-in PubMed, GEO, and PMC providers. This ...
- [Maintaining Documentation - Wiki Maintenance Guide](/docs/developer/maintaining-docs): This guide explains how to maintain the Lobster AI wiki documentation using the automated quality systems.
- [Developer Overview - Lobster AI Architecture](/docs/developer/overview): This guide provides a comprehensive introduction to developing within the Lobster AI codebase, covering architecture patterns, design principles, and develop...
- [Testing Guide - Lobster AI Testing Framework](/docs/developer/testing): This guide provides comprehensive documentation for the Lobster AI testing framework, targeting 95%+ code coverage with scientifically accurate testing scena...
- [AQUADIF Tool Taxonomy](/docs/extending/aquadif-taxonomy): 10-category taxonomy for classifying Lobster AI tools — making the system introspectable, enforceable, and teachable
- [Entry Points](/docs/extending/entry-points): Plugin discovery using Python entry points
- [Extending Lobster](/docs/extending): Create custom agent packages for Lobster AI
- [Package Structure](/docs/extending/package-structure): Standard layout for lobster agent packages
- [Plugin Contract](/docs/extending/plugin-contract): API contract for agent packages in the Lobster ecosystem
- [Scaffold Generator](/docs/extending/scaffold): Generate complete agent packages from the CLI with lobster scaffold agent
- [Testing Agents](/docs/extending/testing): Test patterns for custom agent packages
- [Configuration Guide](/docs/getting-started/configuration): This guide covers all aspects of configuring Lobster AI, from basic API key setup to advanced model customization and cloud integration.
- [Getting Started with Lobster AI](/docs/getting-started): Welcome to Lobster AI! This quick start guide will have you analyzing bioinformatics data in minutes.
- [Comprehensive Installation Guide](/docs/getting-started/installation): This guide covers all installation methods for Lobster AI, from quick setup to advanced development configurations.
- [CLI Commands Reference](/docs/guides/cli-commands): Lobster AI provides a rich command-line interface with enhanced features including Tab completion, command history, and context-aware suggestions. The CLI su...
- [Data Analysis Workflows](/docs/guides/data-analysis-workflows): This guide provides step-by-step workflows for analyzing different types of biological data using Lobster AI. Each workflow combines natural language interac...
- [Data Formats Guide](/docs/guides/data-formats): Lobster AI supports a wide range of biological data formats for different omics types. This guide provides detailed specifications for supported input and ou...
- [Release Notes & Migration Guides](/docs/guides/migration): This document provides comprehensive release notes for Lobster AI versions, covering new features, breaking changes, and recommended upgrade paths for future...
- [Optional Dependencies Guide](/docs/guides/optional-dependencies): This guide covers optional software components that enhance Lobster AI with specialized capabilities. None of these are required for basic functionality, but...
- [User Guide Overview](/docs/guides/overview): Lobster AI is a multi-agent bioinformatics analysis platform that combines specialized AI agents with proven scientific tools to analyze complex multi-om...
- [Semantic Search & Ontology Matching](/docs/guides/semantic-search): Match biomedical terms to standardized ontology concepts using vector embeddings
- [Frequently Asked Questions (FAQ)](/docs/support/faq): This comprehensive FAQ addresses common questions about Lobster AI, covering everything from basic usage to advanced features and troubleshooting.
- [Glossary](/docs/support/glossary): This comprehensive glossary defines bioinformatics, technical, and Lobster AI-specific terms used throughout the documentation and platform.
- [Support](/docs/support): FAQ, troubleshooting, and glossary
- [Troubleshooting Guide](/docs/support/troubleshooting): This comprehensive troubleshooting guide provides solutions to common issues encountered while using Lobster AI. Each problem includes symptoms, causes, and ...
- [Bulk RNA-seq Analysis Tutorial](/docs/tutorials/bulk-rnaseq): This comprehensive tutorial demonstrates how to perform bulk RNA-seq differential expression analysis using Lobster AI with pyDESeq2 integration, formula-bas...
- [Claude Code Integration](/docs/tutorials/claude-code-integration): Lobster AI integrates with Claude Code as a custom skill, enabling bioinformatics analyses directly from your IDE through natural l...
- [Examples Cookbook](/docs/tutorials/cookbook): This comprehensive cookbook provides practical code snippets, analysis recipes, and real-world solutions for common bioinformatics tasks using Lobster AI. Ea...
- [Creating Custom Agents Tutorial](/docs/tutorials/custom-agent): This comprehensive tutorial demonstrates how to create, integrate, and deploy custom AI agents in Lobster AI using the centralized agent registry system and ...
- [Tutorials](/docs/tutorials): Step-by-step analysis tutorials with Lobster AI
- [Proteomics Analysis Tutorial](/docs/tutorials/proteomics): This comprehensive tutorial demonstrates how to analyze both mass spectrometry and affinity proteomics data using Lobster AI's specialized proteomics platfor...
- [Single-Cell RNA-seq Analysis Tutorial](/docs/tutorials/single-cell-rnaseq): This comprehensive tutorial demonstrates how to perform complete single-cell RNA-seq analysis using Lobster AI, from data acquisition to biological interpret...
- [Context](/docs/cloud/assistant/context): How the assistant understands your data and selections
- [Memories](/docs/cloud/assistant/memories): How the assistant retains context across sessions
- [Surface](/docs/cloud/assistant/surface): Where the assistant appears across Omics-OS Cloud
- [Editor](/docs/cloud/editor): LaTeX report editor with AI writing assistant
- [AI Agent Selection](/docs/cloud/features/agent-selection): Toggle specialist AI agents for your analysis session.
- [Agent Store](/docs/cloud/features/agent-store): Browse and enable curated and community AI agents.
- [Canvas to Chat](/docs/cloud/features/canvas-selection): Select data points on plots or tables and ask Lobster about them.
- [Canvas Templates](/docs/cloud/features/canvas-templates): Save and reuse analysis layouts as starter templates for your team.
- [Reasoning & Review](/docs/cloud/features/chain-of-thought): See Lobster's reasoning, tool calls, and scientific review verdicts.
- [OmniBar](/docs/cloud/features/command-bar): Context-aware command surface — search databases, ask the assistant, query tables, navigate the canvas, and capture findings from a single interface
- [Data Upload & Management](/docs/cloud/features/data-management): Upload, browse, and manage multi-omics data files in your workspace.
- [Biological Database Search](/docs/cloud/features/database-search): Search 8 major biological databases from the command bar — UniProt, NCBI Gene, PDB, PubMed, ChEMBL, GEO, KEGG, and NCBI Nucleotide
- [Deep Work Mode](/docs/cloud/features/deep-work-mode): Expand the chat to a full sidebar for focused analysis sessions.
- [Export Your Work](/docs/cloud/features/export): Export analysis as PDF reports, Jupyter notebooks, or slide decks.
- [Fork & Continue](/docs/cloud/features/fork-and-continue): Create editable copies of analysis sessions for independent exploration.
- [Analysis Packages](/docs/cloud/features/org-packages): Publish complete analyses — data, plots, and canvas — for your team.
- [Pathway Exploration](/docs/cloud/features/pathway-exploration): Load KEGG pathways, interact with entities, and resolve genes and compounds to 3D structures and data cards
- [Session Sharing](/docs/cloud/features/session-sharing): Share read-only links to your analysis sessions with collaborators.
- [Slash Commands](/docs/cloud/features/slash-commands): Quick actions and shortcuts available in the chat composer.
- [Workspace](/docs/cloud/features/workspace): Canvas-based scientific workspace — chat with Lobster, build knowledge graphs, manage data, and visualize results
- [Agent Playground](/docs/cloud/playground): Build and test custom agents in the browser
- [Cloud CLI](/docs/cloud/user-guide/cloud-cli): Install and use the Omics-OS Cloud CLI (@omicsos/lobster) to access cloud agents, sessions, files, and vector search from your terminal — no Python required.
- [Getting Started](/docs/cloud/user-guide/getting-started): Start analyzing bioinformatics data with Omics-OS Cloud — sign up, explore the cockpit, and run your first analysis