MCP IPFS Server (storacha.network)

The MCP IPFS Server is a Node.js‑based implementation of the Model Context Protocol that bridges AI assistants with the Storacha network. By wrapping the command‑line interface, it exposes a rich set of tools that allow language models and other MCP clients to manage storage spaces, upload and retrieve data, handle delegations, and monitor usage—all without needing to understand the underlying IPFS commands. This abstraction is especially valuable for developers building AI‑driven applications that require persistent, decentralized storage but prefer a high‑level API over direct command‑line interactions.

At its core, the server provides MCP tools that mirror the most common operations. Authentication and agent management (, , ) let the model verify its identity, while space‑management commands (, , , , ) enable dynamic creation and selection of storage namespaces. Data handling tools (, , ) support uploading, listing, and deleting files or CAR archives. Sharing is simplified through , which generates a public w3s.link URL. Delegation and proof tools (, , etc.) allow fine‑grained access control, and key/token utilities (, ) provide secure credential management. Advanced storage features () expose low‑level operations such as blob uploads, index management, and Filecoin metadata. Finally, billing tools (, , ) give models visibility into costs and quotas.

Developers can integrate the server into any MCP‑compatible workflow—whether through a simple invocation for local testing or via Docker for production deployments. Once connected, an AI assistant can issue high‑level commands like “upload this dataset to space X” or “generate a sharing link for the latest backup,” and the server will translate those into precise calls. This seamless integration removes the need for manual CLI usage, reduces boilerplate code, and ensures consistent error handling across different environments.

Typical use cases include:

• Data‑centric AI services that need to persist training data or model checkpoints on IPFS.
• Decentralized content platforms where an assistant can publish, update, or revoke access to media files.
• Collaborative research environments that require secure delegation of storage rights among team members.
• Cost‑aware applications that monitor storage usage and trigger alerts when quotas approach limits.

The server’s standout advantages are its complete coverage of the CLI, automatic handling of authentication tokens, and support for both high‑level and low‑level operations. By packaging these capabilities behind the MCP interface, it enables AI assistants to treat IPFS storage as a first‑class resource—making decentralized persistence accessible, reliable, and developer‑friendly.