功能描述

Analyze biological data using Lobster AI — single-cell RNA-seq, bulk RNA-seq, literature mining, dataset discovery, quality control, and visualization. USE THIS SKILL WHEN: - Analyzing single-cell or bulk RNA-seq data - Searching PubMed/GEO for papers or datasets - Running quality control on biological data - Clustering cells, finding markers, differential expression - Creating publication-quality visualizations - Working with H5AD, CSV, 10X, GEO/SRA accessions TRIGGER PHRASES: "analyze cells", "search PubMed", "download GEO", "run QC", "cluster", "find markers", "differential expression", "UMAP", "volcano plot", "single-cell", "RNA-seq", "bioinformatics" ASSUMES: Lobster is installed and configured. For setup issues, tell user to run `lobster config-test` and fix any errors before proceeding.

使用说明 (SKILL.md)

Lobster AI Usage Guide

Name: LobsterBio - Use
Author: cewinharhar

Lobster AI is a multi-agent bioinformatics platform. You interact via natural language or slash commands — Lobster routes to specialist agents automatically.

Installation

If Lobster is not installed, guide the user to the right command for their platform:

macOS / Linux

curl -fsSL https://install.lobsterbio.com | bash

Windows (PowerShell)

irm https://install.lobsterbio.com/windows | iex

Manual install (any platform)

uv tool install 'lobster-ai[full,anthropic]' && lobster init
# or: pip install 'lobster-ai[full]' && lobster init

After install, lobster init configures API keys and selects agent packages.

Upgrading

uv tool: uv tool upgrade lobster-ai
pip: pip install --upgrade lobster-ai

Adding Agents (uv tool installs)

Users with uv tool installs add agents via: uv tool install lobster-ai --with lobster-transcriptomics --with lobster-proteomics Running lobster init will guide this process and generate the command.

Quick Reference

Task	Reference
CLI commands	references/cli-commands.md
Single-cell analysis	references/single-cell-workflow.md
Bulk RNA-seq analysis	references/bulk-rnaseq-workflow.md
Literature & datasets	references/research-workflow.md
Visualization	references/visualization.md
Available agents	references/agents.md

Interaction Modes

Interactive Chat

lobster chat                          # Start interactive session
lobster chat --workspace ./myproject  # Custom workspace
lobster chat --reasoning              # Enable detailed reasoning

Single Query

lobster query "Your request"
lobster query --session-id latest "Follow-up request"

Core Patterns

Natural Language (Primary)

Just describe what you want:

"Download GSE109564 and run quality control"
"Cluster the cells and find marker genes"
"Compare hepatocytes vs stellate cells"

Slash Commands (System Operations)

/data                    # Show loaded data info
/files                   # List workspace files
/workspace list          # List available datasets
/workspace load 1        # Load dataset by index
/plots                   # Show generated visualizations
/save                    # Save current session
/status                  # Show system status
/help                    # All commands

Session Continuity

# Start named session
lobster query --session-id "my_analysis" "Load GSE109564"

# Continue with context
lobster query --session-id latest "Now cluster the cells"
lobster query --session-id latest "Find markers for cluster 3"

Agent System

Lobster routes to specialist agents automatically:

Agent	Handles
Supervisor	Routes queries, coordinates agents
Research Agent	PubMed search, GEO discovery, paper extraction
Data Expert	File loading, format conversion, downloads
Transcriptomics Expert	scRNA-seq: QC, clustering, markers
DE Analysis Expert	Differential expression, statistical testing
Annotation Expert	Cell type annotation, gene set enrichment
Visualization Expert	UMAP, heatmaps, volcano plots
Proteomics Expert	Mass spec analysis [alpha]
Genomics Expert	VCF, GWAS, variant analysis [alpha]
ML Expert	Embeddings, classification [alpha]

Workspace & Outputs

Default workspace: .lobster_workspace/

Output files:

Extension	Content
`.h5ad`	Processed AnnData objects
`.html`	Interactive visualizations
`.png`	Publication-ready plots
`.csv`	Exported tables
`.json`	Metadata, provenance

Managing outputs:

/files              # List all outputs
/plots              # View visualizations
/open results.html  # Open in browser
/read summary.csv   # Preview file contents

Typical Workflow

# 1. Start session
lobster chat --workspace ./my_analysis

# 2. Load or download data
"Download GSE109564 from GEO"
# or
/workspace load my_data.h5ad

# 3. Quality control
"Run quality control and show me the metrics"

# 4. Analysis
"Filter low-quality cells, normalize, and cluster"

# 5. Interpretation
"Identify cell types and find marker genes"

# 6. Visualization
"Create UMAP colored by cell type"
/plots

# 7. Export
"Export marker genes to CSV"
/save

Troubleshooting Quick Reference

Issue	Check
Lobster not responding	`lobster config-test`
No data loaded	`/data` to verify, `/workspace list` to see available
Analysis fails	Try with `--reasoning` flag
Missing outputs	Check `/files` and workspace directory

Documentation

Online docs: docs.omics-os.com

Key sections:

Guides → CLI Commands
Guides → Data Analysis Workflows
Tutorials → Single-Cell RNA-seq
Agents → Package documentation

安全使用建议

This skill appears to be what it says (a usage guide for Lobster AI) but has two practical concerns you should address before proceeding: 1) Installer provenance: the SKILL.md recommends running curl | bash from https://install.lobsterbio.com and a Windows 'irm ... | iex' — running scripts directly from an unfamiliar domain can execute arbitrary code on your machine. Prefer installing from a known package source (pip or your OS package manager) or inspect the installer contents on a trusted network before running. If possible, ask for the project's public repository or official release page to verify the installer. 2) Hidden credential needs: the docs reference an 'anthropic' extras tag and say 'lobster init' configures API keys. The skill metadata does not declare required API keys or where credentials will be stored. Before use, verify what external services (model vendors, PubMed/GEO access, cloud storage) Lobster will contact and what keys it needs; run 'lobster config-test' and read any config files to see where secrets are kept. Additional steps: request the skill's source/homepage from the publisher (none provided), prefer installing via pip or a known package repository, run installations in a sandbox or VM first, and review the installer script/ repository for network endpoints or telemetry before granting access to sensitive data. If you need higher assurance, ask the skill author for a link to an official repo or signed releases.

功能分析

Type: OpenClaw Skill Name: lobster-bio-use Version: 1.0.1 The skill bundle is classified as suspicious due to high-risk installation methods and broad file system access capabilities that present significant vulnerabilities if the OpenClaw agent is susceptible to prompt injection or lacks proper sandboxing. Specifically, `SKILL.md` and `references/cli-commands.md` document `curl -fsSL https://install.lobsterbio.com | bash` and `irm https://install.lobsterbio.com/windows | iex` for installation, which are direct remote code execution patterns. Additionally, `references/cli-commands.md` details commands like `/read <file>` and `/open <file>` with pattern support, which could be exploited to read arbitrary sensitive files or execute arbitrary code if the agent is prompted to do so, without explicit restrictions to its own generated outputs. There is no clear evidence of intentional malicious behavior within the provided files, but the described capabilities pose a high risk of exploitation.

能力评估

✓ Purpose & Capability

Name/description align with the runtime instructions and reference files: all content is about using Lobster AI for single-cell/bulk RNA‑seq, literature/dataset discovery, QC, DE, and visualization. The required capabilities described are consistent with the workflows and agent list in the references.

ℹ Instruction Scope

SKILL.md is instruction-only and stays within Lobster usage: commands, workspace paths, and dataset downloads (GEO/SRA) are all relevant. It assumes Lobster is installed/configured and instructs user to run 'lobster config-test' for setup issues. The instructions do not ask the agent to read unrelated system files or undeclared env vars, but they do direct users to download remote datasets and to run an external installer (see install_mechanism).

⚠ Install Mechanism

The guide recommends piping a shell download from https://install.lobsterbio.com (curl | bash) and a PowerShell 'irm ... | iex' for Windows — both are high-risk patterns because they execute code directly from a remote, non-standard domain. There are fallback install options (pip, uv tool), but provenance for the install.lobsterbio.com endpoint is unknown and no homepage/repository is provided in the skill metadata. This is the primary risk.

ℹ Credentials

The skill declares no required env vars or primary credential, which matches that it is instruction-only. However, installation guidance references packages with an 'anthropic' extras marker and says 'lobster init configures API keys' — implying third-party model/API keys (Anthropic or others) may be created/required. The skill does not declare or document these credential needs or where keys are stored, which is a proportionality/provenance gap to be aware of.

✓ Persistence & Privilege

The skill is instruction-only, does not request 'always: true', and does not modify other skills or system-wide agent settings. It does not ask to persist additional privileges itself.

版本历史

v1.0.1

- Added detailed installation and upgrade instructions for Lobster AI across all platforms, including uv tool, pip, and PowerShell methods. - Provided guidance for adding specialist agent packages using uv tool and lobster init. - Clarified agent availability, indicating Proteomics, Genomics, and ML agents as [alpha] rather than [premium]. - No code or logic changes; documentation and usage guide expanded to improve onboarding and setup clarity.

v1.0.0

- First public release of the lobster-use skill for biological data analysis. - Supports natural language and slash-command workflows for single-cell/bulk RNA-seq, literature mining, and visualization. - Integrates PubMed/GEO search, dataset loading, comprehensive quality control, clustering, marker finding, and differential expression. - Automatic routing to specialized agents for research, data management, analysis, visualization, and more. - Provides concise quick-reference guides, example workflows, and troubleshooting resources. - Assumes Lobster is installed and configured; directs users to run `lobster config-test` for setup issues.

元数据

Slug lobster-bio-use

版本 1.0.1

许可证 —

累计安装 2

当前安装数 2

历史版本数 2

常见问题

LobsterBio - Use 是什么？

Analyze biological data using Lobster AI — single-cell RNA-seq, bulk RNA-seq, literature mining, dataset discovery, quality control, and visualization. USE THIS SKILL WHEN: - Analyzing single-cell or bulk RNA-seq data - Searching PubMed/GEO for papers or datasets - Running quality control on biological data - Clustering cells, finding markers, differential expression - Creating publication-quality visualizations - Working with H5AD, CSV, 10X, GEO/SRA accessions TRIGGER PHRASES: "analyze cells", "search PubMed", "download GEO", "run QC", "cluster", "find markers", "differential expression", "UMAP", "volcano plot", "single-cell", "RNA-seq", "bioinformatics" ASSUMES: Lobster is installed and configured. For setup issues, tell user to run `lobster config-test` and fix any errors before proceeding. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 1276 次。

如何安装 LobsterBio - Use？

在 OpenClaw 或 Claude Code 对话框中运行命令「/install lobster-bio-use」即可一键安装，无需额外配置。

LobsterBio - Use 是免费的吗？

是的，LobsterBio - Use 完全免费（开源免费），可自由下载、安装和使用。

LobsterBio - Use 支持哪些平台？

LobsterBio - Use 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。

谁开发了 LobsterBio - Use？

由 cewinharhar（@cewinharhar）开发并维护，当前版本 v1.0.1。

LobsterBio - Use