该市场为生命科学工具提供MCP（模型上下文协议）服务器和技能。安装这些插件即可直接访问Claude代码中的专业研究和分析工具。

包含内容：

• MCP 服务器：连接外部服务如 PubMed、BioRender、Synapse 等
• 技能：扩展Claude专业知识的领域特定工作流程和分析能力

快速启动

对于需要认证的服务器（除PubMed外），安装后请配置凭证：

1. 在 Claude Code 中类型 /plugin
2. 选择“管理插件”
3. 找到你安装的服务器
4. 选择“配置”
5. 输入所需凭证
6. 重启 Claude 代码

可用插件

远程MCP服务器

PubMed

插件ID：pubmed@life-sciences

搜索并访问PubMed上的生物医学文献和研究文章。

需求：无——所有用户均可访问

生物渲染

插件ID：biorender@life-sciences

创建并访问科学插图和图表。

要求：免费BioRender账户（https://www.biorender.com)

####Synapse.org 插件ID：synapse@life-sciences

Life Sciences Marketplace for Claude Code

This marketplace provides skills for life sciences research. Skills are domain-specific workflows that extend Claude’s expertise.

Quick Start

# Add the marketplace
/plugin marketplace add anthropics/life-sciences

# Install skills
/plugin install clinical-trial-protocol@life-sciences
/plugin install scientific-problem-selection@life-sciences
/plugin install binder-gauntlet@life-sciences

Available Skills

Clinical Trial Protocol

Plugin ID: clinical-trial-protocol@life-sciences

Generate FDA/NIH-compliant clinical trial protocols for medical devices or drugs.

Scientific Problem Selection

Plugin ID: scientific-problem-selection@life-sciences

Systematic framework for scientific problem selection and strategic research decisions. Based on Fischbach & Walsh’s methodology from Cell (2024), this skill helps researchers with project ideation, risk assessment, troubleshooting stuck projects, and strategic scientific planning.

Use cases:

• Pitch and refine new research ideas
• Evaluate project risks and feasibility
• Navigate decision trees in active projects
• Strategic research planning and problem choice

Clinical Trial Forecast

Plugin ID: clinical-trial-forecast@life-sciences

Estimate the probability that a registered clinical trial meets its goal, with a cited report and the simulation model behind the number. Runs in Claude Science; see the skill’s README.

Binder Gauntlet

Plugin ID: binder-gauntlet@life-sciences

Validate de novo designed protein binders with independent re-prediction and controls, profile their developability, and write a dossier that states what was not checked. All results are computational predictions.

Removed Plugins

This repository has been refocused on skills, so it no longer carries MCP connector plugins. If you installed one, it will no longer update from this marketplace. Earlier versions remain available in the git history.

The following skills were also removed: instrument-data-to-allotrope, nextflow-development, scvi-tools and single-cell-rna-qc.