Storacha MCP Storage Server

The Model Context Protocol (MCP) server implementation for Storacha provides a standardized interface for AI applications to store and retrieve files. It enables trustless, decentralized data exchange using IPFS and Content Identifiers (CIDs), ensuring data sovereignty and verifiability.

Free Storage Options

Start using Storacha with these generous free storage options:

GitHub Users: Get 100MB free storage instantly when signing up with GitHub - no credit card required
Email Users: Receive 5GB free storage when registering with email and adding a credit card

Setting Up the MCP Server

Installation

Clone the repository and install dependencies:

# Clone the repository
git clone https://github.com/storacha/mcp-storage-server.git
cd mcp-storage-server
 
# Install dependencies
pnpm install

Generate Keys & Create Delegation

First, install the CLI tool:

npm install -g @web3-storage/w3cli

Then login to Storacha:

w3 login

Select Via GitHub (or Email) and follow returned steps to authenticate.

Create a space for your data:

w3 space create your-space-name

⚠️

Save the recovery key to access your space from other devices.

Generate a private key:

w3 key create

This outputs an AgentId and PrivateKey that you'll need in the next step.

Create a delegation with proper permissions:

w3 delegation create your-agent-id \
  --can 'filecoin/offer' \
  --can 'upload/add' \
  --can 'space/blob/add' \
  --can 'space/index/add' --base64

Replace your-agent-id with the AgentId from the previous step.

Configure Your MCP Client

Set up your MCP client to use the Storacha server by adding this configuration:

{
  "mcpServers": {
    "storacha-storage-server": {
      "command": "node",
      "args": ["./dist/index.js"],
      "env": {
        "MCP_TRANSPORT_MODE": "stdio",
        "PRIVATE_KEY": "your-agent-private-key",
        "DELEGATION": "your-base64-delegation"
      },
      "shell": true,
      "cwd": "./"
    }
  }
}

Replace the placeholder values with your actual PrivateKey and base64 delegation.

Using the MCP Tools

Storacha MCP Server provides the following tools for interacting with decentralized storage.

Upload Tool

Upload files to the Storacha Network:

// Example usage in an AI application
const result = await uploadFile({
  file: base64EncodedContent,
  name: "document.pdf",
  publishToFilecoin: true
});

Parameters:

file: Base64 encoded file content
name: Filename with extension
publishToFilecoin: (Optional) Whether to publish to Filecoin
delegation: (Optional) Custom delegation proof
gatewayUrl: (Optional) Custom gateway URL

Retrieve Tool

Retrieve files from the Storacha Network:

// Example retrieval by CID
const document = await retrieveFile({
  filepath: "bafybei...gq5a/document.pdf"
});

Parameters:

filepath: CID path in format: CID/filename, /ipfs/CID/filename, or ipfs://CID/filename
useMultiformatBase64: (Optional) Whether to use multiformat base64 encoding

Identity Tool

Retrieve the DIDKey of the Storacha Agent:

const identity = await getIdentity({});

Use Cases

Document Storage & Analysis: Store and retrieve documents for AI processing
Structured Data Storage: Maintain data optimized for longevity and accessibility
Cross-Agent Data Sharing: Share data between AI agents using CIDs
AI Model Development: Support AI models with reliable access to datasets
LLM Integration: Connect LLMs directly with Storacha for data access

Advanced Integration

For more detailed integration patterns including SDK usage and Docker deployment, refer to the integrations guide (opens in a new tab).

Support

For support, join our Storacha Discord sever (opens in a new tab) or open an issue in the GitHub repository.

ElizaOS Create an account