biostratum-eliza

Biostratum MCP Plugin for ElizaOS

Developed during Bio x AI Hackathon 2025 as part of HolyBioMCP project

Bio-Gagarin character with biostratum plugin working in ElizaOS native chat

This plugin integrates the Model Context Protocol (MCP) with ElizaOS, allowing agents to connect to multiple MCP servers and use their resources, prompts, and tools. It includes specialized support for biostratum servers that provide access to biological and longevity research data.

🔍 What is MCP?

The Model Context Protocol (MCP) is an open protocol that enables seamless integration between LLM applications and external data sources and tools. It provides a standardized way to connect LLMs with the context they need.

This plugin allows your ElizaOS agents to access multiple MCP servers simultaneously, each providing different capabilities:

• Resources: Context and data for the agent to reference
• Prompts: Templated messages and workflows
• Tools: Functions for the agent to execute

📦 Installation

Option 1: ElizaOS v1.0+ (Recommended)

Use the ElizaOS CLI for the simplest setup:

# Install the plugin
elizaos plugins add @longevity-genie/biostratum-mcp-plugin

# Start your agent with a character configuration
elizaos agent start --path bio-gagarin.json

Option 2: Traditional Package Manager Installation

Install the plugin directly in your ElizaOS project:

• npm

npm install @longevity-genie/biostratum-mcp-plugin

• pnpm

pnpm install @longevity-genie/biostratum-mcp-plugin

• yarn

yarn add @longevity-genie/biostratum-mcp-plugin

• bun

bun add @longevity-genie/biostratum-mcp-plugin

🚀 Usage

Option 1: Traditional MCP Configuration

Add the plugin to your character configuration:

{
  "name": "Your Character",
  "plugins": ["@longevity-genie/biostratum-mcp-plugin"],
  "settings": {
    "mcp": {
      "servers": {
        "github": {
          "type": "stdio",
          "name": "Code Server",
          "command": "npx",
          "args": ["-y", "@modelcontextprotocol/server-github"]
        }
      }
    }
  }
}

Option 2: Biostratum Configuration (Simplified)

For easier setup with biostratum-compatible servers:

{
  "name": "Your Character",
  "plugins": ["@longevity-genie/biostratum-mcp-plugin"],
  "settings": {
    "biostratum": {
      "biothings": {},
      "opengenes": {},
      "gget": {},
      "pharmacology": {},
      "synergy-age": {}
    }
  }
}

This automatically configures biostratum MCP servers using uvx. All five ready servers (biothings, opengenes, gget, pharmacology, synergy-age) are enabled by default with all their tools available. You can also mix both configurations - biostratum and mcp servers will be merged.

📦 Prerequisites for Biostratum Servers

To use biostratum servers (like biothings), you need to install uv:

# Download and install uv
curl -LsSf https://astral.sh/uv/install.sh | sh

After installation, restart your terminal or run:

source ~/.bashrc  # or ~/.zshrc depending on your shell

⚙️ Configuration Options

MCP supports two types of servers: "stdio" and "sse". Each type has its own configuration options.

Common Options

Option	Type	Description
type	string	The type of MCP server: "stdio" or "sse"
name	string	The display name of the server
timeout	number	Timeout in seconds for tool calls (default: 60)
disabled	boolean	Whether the server is disabled

stdio Server Options

Option	Type	Description
command	string	The command to run the MCP server
args	string[]	Command-line arguments for the server
env	object	Environment variables to pass to the server
cwd	string	Optional Working directory to run the server in

sse Server Options

Option	Type	Description
url	string	The URL of the SSE endpoint

Biostratum Configuration

The biostratum section provides simplified configuration for biostratum-compatible servers. Each server key acts as an on/off switch:

{
  "settings": {
    "biostratum": {
      "biothings": {
        // Optional: override default stdio configuration
        "enabled": true,
        "include": ["query_genes", "get_gene", "query_variants", "get_variant"], // Only these tools
        "exclude": ["download_entrez_data", "perform_pairwise_alignment"], // Exclude these tools
        "env": {
          "CUSTOM_VAR": "value"
        },
        "timeout": 120
      },
      "gget": {
        "include": ["gget_search", "gget_info", "gget_seq", "gget_blast", "gget_alphafold"] // Popular genomics tools
      },
      "synergy-age": {
        "include": ["synergyage_get_schema_info", "synergyage_example_queries"] // Schema and examples only
      },
      "longevity": {
        "enabled": false  // WIP server, disabled by default
      }
    }
  }
}

Available Biostratum Servers

Server Name	Status	Tools	Description	Default Command
biothings	✅ Ready	19	Access to biological data and annotations (genes, variants, chemicals, taxa)	uvx biothings-mcp
opengenes	✅ Ready	3	Gene information related to longevity and aging research	uvx opengenes-mcp
gget	✅ Ready	19	Comprehensive genomics toolkit (search, sequences, BLAST, structure prediction)	uvx gget-mcp
pharmacology	✅ Ready	7	Pharmacological targets, ligands, and interactions from Guide to PHARMACOLOGY	uvx pharmacology-mcp stdio
longevity	🚧 WIP	-	Longevity research data and interventions	uvx longevity-mcp
synergy-age	✅ Ready	3	Database of synergistic and antagonistic lifespan effects of genetic interventions in model organisms	uvx synergy-age-mcp
druginteractions	🚧 WIP	-	Drug interaction data and analysis	uvx druginteractions-mcp

Note:

• ✅ Ready servers are enabled by default and fully functional (total: 51 tools)
• 🚧 WIP servers are disabled by default and still in development. Set "enabled": true to use them.

Detailed Tool Breakdown

Biothings Server (19 tools):

• Gene tools: query_genes, query_many_genes, get_gene, get_genes, get_gene_metadata
• Variant tools: query_variants, query_many_variants, get_variant, get_variants
• Chemical tools: query_chems, query_many_chemicals, get_chem, get_chems
• Taxonomy tools: get_taxon, get_taxons, query_taxons, query_many_taxons
• Utility tools: download_entrez_data, perform_pairwise_alignment

OpenGenes Server (3 tools):

• opengenes_get_schema_info: Get database schema information
• opengenes_example_queries: Get example SQL queries
• opengenes_db_query: Query longevity and aging gene database

Gget Server (19 tools):

• Search & Info: gget_search, gget_info, gget_ref
• Sequences: gget_seq, gget_blast, gget_blat, gget_muscle, gget_diamond
• Expression: gget_archs4, gget_cellxgene
• Function: gget_enrichr, gget_elm, gget_bgee
• Structure: gget_alphafold, gget_pdb
• Disease/Drugs: gget_cosmic, gget_opentargets
• Utilities: gget_mutate, gget_setup

Pharmacology Server (7 tools):

• Core API: list_targets, list_ligands, list_interactions
• File Output: local_search_targets_to_file, local_search_ligands_to_file, local_get_target_interactions_to_file, local_get_ligand_interactions_to_file

SynergyAge Server (3 tools):

• synergyage_get_schema_info: Get database schema information for genetic synergy data
• synergyage_example_queries: Get example SQL queries for synergy/epistasis analysis
• synergyage_db_query: Query database of synergistic and antagonistic lifespan effects from genetic interventions

The SynergyAge database contains curated information about synergistic and antagonistic lifespan effects:

• 1,790 mutants with at least 2 genetic interventions across 3 model organisms
• 589 unique genes with 6,967 lifespan values from peer-reviewed literature
• C. elegans: 1,727 gene combinations with 6,635 lifespan values
• D. melanogaster: 27 gene combinations with 185 lifespan values
• M. musculus: 36 gene combinations with 147 lifespan values
• Various intervention types: mutations, RNA interference, overexpression
• Helps determine if lifespan determinant effects are additive or involve complex genetic interactions

Each server supports multiple running modes:

• STDIO Mode (default): uvx <server-name>-mcp - Used by this plugin
• HTTP Mode: uvx <server-name>-mcp server --port 8000
• SSE Mode: uvx <server-name>-mcp sse

Tool Filtering

You can control which tools are available from each server using include/exclude patterns:

Option	Type	Description	Default
include	string[]	Only include tools with these exact names	[] (include all)
exclude	string[]	Exclude tools with these exact names	[] (exclude none)

Filtering Logic:

1. If no include/exclude options are specified → all server tools are available
2. If include is specified and not empty → only include tools with matching names
3. If exclude is specified and not empty → exclude tools with matching names (applied after include)
4. Missing/mismatched tool names don't cause errors

Example configurations:

{
  "biothings": {
    "include": ["query_genes", "get_gene", "query_variants", "get_variant"]  // Only gene and variant tools
  },
  "opengenes": {
    "exclude": ["opengenes_db_query"]  // All tools except direct SQL queries
  },
  "gget": {
    "include": ["gget_search", "gget_info", "gget_seq", "gget_alphafold"]  // Core genomics workflow tools
  },
  "pharmacology": {
    "include": ["list_targets", "list_ligands", "list_interactions"]  // Core pharmacology API tools
  },
  "synergy-age": {
    "include": ["synergyage_get_schema_info", "synergyage_example_queries"]  // Schema and examples only
  }
}

🛠️ Using MCP Capabilities

Once configured, the plugin automatically exposes MCP servers' capabilities to your agent:

Context Providers

The plugin includes three providers that add MCP capabilities to the agent's context:

1. MCP_SERVERS: Lists available servers and their tools, resources and prompts

🔄 Plugin Flow

The following diagram illustrates the MCP plugin's flow for tool selection and execution:

graph TD
    %% Starting point - User request
    start[User Request] --> action[CALL_TOOL Action]

    %% MCP Server Validation
    action --> check{MCP Servers Available?}
    check -->|No| fail[Return No Tools Available]
    
    %% Tool Selection Flow
    check -->|Yes| state[Get MCP Provider Data]
    state --> prompt[Create Tool Selection Prompt]
    
    %% First Model Use - Tool Selection
    prompt --> model1[Use Language Model for Tool Selection]
    model1 --> parse[Parse Selection]
    parse --> retry{Valid Selection?}
    
    %% Second Model Use - Retry Selection
    retry -->|No| feedback[Generate Feedback]
    feedback --> model2[Use Language Model for Retry]
    model2 --> parse
    
    %% Tool Selection Result
    retry -->|Yes| toolAvailable{Tool Available?}
    toolAvailable -->|No| fallback[Fallback Response]
    
    %% Tool Execution Flow
    toolAvailable -->|Yes| callTool[Call MCP Tool]
    callTool --> processResult[Process Tool Result]
    
    %% Memory Creation
    processResult --> createMemory[Create Memory Record]
    createMemory --> reasoningPrompt[Create Reasoning Prompt]
    
    %% Third Model Use - Response Generation
    reasoningPrompt --> model3[Use Language Model for Response]
    model3 --> respondToUser[Send Response to User]
    
    %% Styling
    classDef model fill:#f9f,stroke:#333,stroke-width:2px;
    classDef decision fill:#bbf,stroke:#333,stroke-width:2px;
    classDef output fill:#bfb,stroke:#333,stroke-width:2px;
    
    class model1,model2,model3 model;
    class check,retry,toolAvailable decision;
    class respondToUser,fallback output;

📋 Example: Setting Up Multiple MCP Servers

Here's a complete example configuration with multiple MCP servers using both traditional MCP configuration and biostratum:

{
  "name": "Developer Assistant",
  "plugins": ["@longevity-genie/biostratum-mcp-plugin", "other-plugins"],
  "settings": {
    "mcp": {
      "servers": {
        "github": {
          "type": "stdio",
          "command": "npx",
          "args": ["-y", "@modelcontextprotocol/server-github"],
          "env": {
            "GITHUB_PERSONAL_ACCESS_TOKEN": "<YOUR_TOKEN>"
          }
        },
        "puppeteer": {
          "type": "stdio",
          "command": "npx",
          "args": ["-y", "@modelcontextprotocol/server-puppeteer"]
        }
      },
      "maxRetries": 2
    },
    "biostratum": {
      "biothings": {
        "include": ["query_genes", "get_gene", "query_variants", "get_variant"],
        "timeout": 120
      },
      "opengenes": {
        "exclude": ["opengenes_db_query"]
      },
      "gget": {
        "include": ["gget_search", "gget_info", "gget_seq", "gget_blast", "gget_alphafold"]
      },
      "pharmacology": {
        "include": ["list_targets", "list_ligands", "list_interactions"]
      },
      "synergy-age": {},
      "longevity": {
        "enabled": true
      }
    }
  }
}

You can also use only biostratum configuration:

{
  "name": "Biomedical Assistant",
  "plugins": ["@longevity-genie/biostratum-mcp-plugin"],
  "settings": {
    "biostratum": {
      "biothings": {
        "exclude": ["download_entrez_data", "perform_pairwise_alignment"]
      },
      "opengenes": {
        "include": ["opengenes_get_schema_info", "opengenes_example_queries"]
      },
      "gget": {
        "include": ["gget_search", "gget_info", "gget_seq", "gget_blast", "gget_alphafold", "gget_cosmic", "gget_opentargets"]
      },
      "synergy-age": {
        "include": ["synergyage_get_schema_info", "synergyage_db_query"]
      }
    }
  }
}

🔒 Security Considerations

Please be aware that MCP servers can execute arbitrary code, so only connect to servers you trust.

🔍 Troubleshooting

If you encounter issues with the MCP plugin:

1. Check that your MCP servers are correctly configured and running
2. Ensure the commands are accessible in the ElizaOS environment
3. Review the logs for connection errors
4. Verify that the plugin is properly loaded in your character configuration

👥 Contributing

Thanks for considering contributing to our project!

How to Contribute

1. Fork the repository.
2. Create a new branch: git checkout -b feature-branch-name.
3. Make your changes.
4. Commit your changes using conventional commits.
5. Push to your fork and submit a pull request.

📄 License

This plugin is released under the same MIT license as ElizaOS.