Filebase (Web3 Storage) MCP. Manage all decentralized assets from a single conversation.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
Filebase (Web3 Storage) provides full control over decentralized storage via IPFS and IPNS. Use this MCP to manage persistent content identifiers, provision network gateways, track usage metrics, and handle keypair generation—all through natural language commands from your AI agent.
What your AI agents can do
Platform create gateway
Creates a new dedicated network gateway endpoint.
Platform create name
Registers and creates a brand new IPNS name record.
Platform delete gateway
Removes an existing dedicated network gateway from the platform.
Check total storage capacity used across buckets or monitor bandwidth usage for specific gateways.
Generate cryptographic keypairs, publish content identifiers (CIDs) to an IPNS name, and resolve those names back into live file paths.
Create, update, list, or delete dedicated network gateways used to access your decentralized storage.
Add, remove, replace, or list content identifiers (CIDs) using the Pinning Service API (PSA) to ensure data remains available long-term.
Upload text files directly to IPFS or retrieve raw blocks and file contents by their CID.
Ask AI about this MCP
Supported MCP Clients
OAuth 2.0 Compatible Gemini Waiting for input…
Filebase (Web3 Storage) - 29 Tools
These tools let you manage every aspect of your Web3 infrastructure: from creating gateways to resolving names, handling the full lifecycle of decentralized data.
Make your AI actually useful.
Add this MCP to Claude, Cursor, or Windsurf and your AI stops guessing. It gets real tools to look things up, take action, and handle the stuff you keep doing by hand.
Start using Filebase (Web3 Storage) on Vinkius019e3896platform create gateway
Creates a new dedicated network gateway endpoint.
019e3896platform create name
Registers and creates a brand new IPNS name record.
019e3896platform delete gateway
Removes an existing dedicated network gateway from the platform.
019e3896platform delete name
Deletes a previously registered IPNS name record.
019e3896platform get bucket usage
Retrieves the specific storage usage details for one designated bucket.
019e3896platform get gateway
Fetches detailed information about a particular dedicated gateway.
019e3896platform get gateway usage
Checks the current bandwidth usage statistics for a specific gateway.
019e3896platform get name
Gets all details related to a specified IPNS name.
019e3896platform get usage
Retrieves the total storage and overall bandwidth usage across the entire system.
019e3896platform list gateways
Returns a list of all active, dedicated network gateways configured.
019e3896platform list names
Lists all currently registered IPNS names in the system.
019e3896platform update gateway
Modifies settings for an already existing dedicated gateway.
019e3896platform update name
Changes an IPNS name to point to a new content identifier (CID).
019e3896psa add pin
Adds a specific CID to the pinning service for guaranteed, long-term availability.
019e3896psa get pin
Checks the status of a pin using its unique request ID.
019e3896psa list pins
Lists all currently active pinning requests via the Pinning Service API.
019e3896psa remove pin
Removes a pin from the service using its request ID.
019e3896psa replace pin
Replaces an existing pinned object with new content while keeping the same identifier.
019e3896rpc add
Uploads and stores a plain text file directly onto the IPFS network.
019e3896rpc block get
Retrieves raw data blocks from IPFS using their specific CID.
019e3896rpc cat
Fetches the readable contents of a file given its CID.
019e3896rpc key gen
Creates and stores a brand new cryptographic keypair in your local keychain.
019e3896rpc key list
Shows all the keypairs currently stored in your managed keychain.
019e3896rpc name publish
Registers a given CID to an IPNS name for public resolution.
019e3896rpc name resolve
Resolves a known IPNS name back into its corresponding, usable IPFS path/CID.
019e3896rpc pin add
Adds an object's CID to persistent storage using the raw RPC method.
019e3896rpc pin ls
Lists all objects currently pinned via the raw RPC interface.
019e3896rpc pin rm
Unpins a content identifier (CID) using the raw RPC method.
019e3896rpc version
Gets the running version number of the IPFS daemon for compatibility checks.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with Filebase (Web3 Storage), then connect any of our 5,000+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 5,000+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Filebase. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Works with Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This server provides 29 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Dealing with decentralized assets means hopping between five different consoles.
Today, managing a single web3 asset is tedious. You upload the file to one dashboard, then copy the CID into another system to check its usage, and finally switch to a third CLI tool just to pin it for permanence. It's constant context switching, copy-pasting identifiers, and waiting for different services to confirm status.
With this MCP, you tell your agent what needs doing—for example, 'Publish the metadata and make sure it stays online.' Your agent handles the key generation, the IPNS publishing (rpc_name_publish), and the pinning service call (psa_add_pin) all in one sequence. You just get confirmation that the task is done.
Filebase (Web3 Storage) MCP: Full Infrastructure Control
The biggest manual step that goes away is auditing resource health. Instead of checking three different panels for storage, bandwidth, and gateway status, you can run platform_list_gateways and then use platform_get_usage to get a single, unified picture.
What's different now is the shift from executing isolated commands to managing an entire lifecycle through natural conversation. You’re treating complex infrastructure like just another step in your workflow.
What you can do with this MCP connector
Connecting Filebase means you're taking the keys to your decentralized infrastructure. Instead of switching between a command line interface or a web dashboard to deal with CIDs, pins, and names, you manage it all directly via conversation. Your agent can generate keypairs for you, publish assets to IPNS, and resolve those names back into usable IPFS paths.
Need to track bandwidth? You just ask your AI client to check the bucket usage or list active gateways. This lets you treat complex Web3 infrastructure management like a simple conversational workflow. Since Vinkius hosts this MCP in its catalog, you get access to all these advanced tools—from raw block retrieval using RPC to managing entire gateway lifecycles—all from one place.
019e3896-7338-7144-b674-921afc602f82 
How Filebase (Web3 Storage) MCP Works
- 1 First, you subscribe to this MCP on Vinkius and enter your Filebase API Key and Platform Token into your AI client.
- 2 Next, you tell your agent the specific infrastructure task. For example, 'List all dedicated gateways' or 'Pin CID QmXoyp...'.
- 3 The tool executes the command against the decentralized network, giving you real-time data like usage metrics, a list of available names, or confirmation that content has been pinned.
The bottom line is, it lets your AI client interact with complex Web3 protocols—like IPFS and IPNS—as if they were simple commands in a chat window.
Who Is Filebase (Web3 Storage) MCP For?
This MCP is for the infrastructure architect who hates switching between terminals. It’s for anyone dealing with decentralized assets whose job requires managing network resources, keypairs, and persistent storage identifiers.
Needs to generate keypairs using rpc_key_gen, publish a new asset CID via rpc_name_publish, and then resolve the final IPFS path for their dApp.
Uses platform_list_gateways to audit existing infrastructure or calls platform_delete_gateway when a service is retired. They also check usage with platform_get_usage.
Manages decentralized metadata by adding text files using rpc_add, then ensures it stays online by calling psa_add_pin for permanent storage.
What Changes When You Connect
- You don't have to leave your chat window. Instead of running multiple CLI commands, you tell your agent to list gateways or check usage with platform_list_gateways and platform_get_usage.
- The system handles complex key management for you. You can use rpc_key_gen to create a pair and rpc_name_publish to register it without ever touching the command line.
- Pinning becomes automatic. When you need content to last forever, you don't just upload it; you call psa_add_pin, ensuring that data remains accessible long after your initial upload expires.
- Full infrastructure control means no guessing games. You can use platform_get_gateway and platform_list_gateways together to map out exactly which network endpoints are active on your account.
- You get granular visibility into resource consumption. Need to know if you're running out of space? Use platform_get_bucket_usage or platform_get_usage; it’s immediate, actionable data.
Real-World Use Cases
Auditing a new decentralized project
A DevOps engineer needs to know if the initial infrastructure setup is complete. They ask their agent to run platform_list_gateways and then call platform_get_usage to ensure both network access points are provisioned and budgeted correctly.
Deploying a mutable website
A content creator wants a decentralized site. They use rpc_key_gen, then call rpc_name_publish to link the new CID via IPNS, and finally ask the agent to resolve it with rpc_name_resolve so they have the final URL.
Securing mission-critical data
A developer uploads a file using rpc_add. To guarantee it never disappears, they immediately tell their agent to run psa_add_pin on that content identifier (CID). This ensures persistent storage.
Debugging resource issues
The ops team notices high bandwidth usage. They use platform_get_gateway_usage for the primary gateway and then check overall capacity with platform_get_usage to isolate if the issue is localized or global.
The Tradeoffs
Assuming file upload is enough
A user uploads a CID using rpc_add and assumes the data will stay online indefinitely. They run their app, but the content disappears after 7 days because it wasn't pinned.
→ Always follow up an initial upload (rpc_add) with a pinning call like psa_add_pin or rpc_pin_add. The pin is what makes it persistent.
Manually managing gateway states
A team member needs to disable a test gateway, but they don't know if it was created via the dashboard, CLI, or an old script.
→ Always start by listing everything. Use platform_list_gateways first, then use platform_delete_gateway on the specific one you need to decommission.
Trying to resolve names without publishing them
A developer runs rpc_name_resolve for a new asset name that was never registered. The operation fails because the IPNS record doesn't exist.
→ Before resolving, you must publish the CID using rpc_name_publish to create the necessary IPNS link.
When It Fits, When It Doesn't
Use this MCP if your workflow involves more than just uploading a file. If you need to manage infrastructure state—like provisioning gateways (platform_create_gateway), tracking complex usage metrics (platform_get_usage), or handling decentralized identifiers (IPNS naming and key pairs)—this is what you want. It lets you automate the entire lifecycle, from creation to retirement. Don't use this if you just need simple file storage; a basic object store connector might be enough. But if your data lives on IPFS, then you gotta use these tools because they handle the complex naming and pinning that keeps things online.
Common Questions About Filebase (Web3 Storage) MCP
How do I list all my current IPNS names using Filebase (Web3 Storage) MCP? +
You use platform_list_names. This tool returns a comprehensive inventory of every IPNS name you've registered, letting you see what assets are publicly addressable via the network.
Do I need to run rpc_key_gen before publishing an asset with Filebase (Web3 Storage) MCP? +
Yes. You must use rpc_key_gen first; this creates the cryptographic keypair needed to sign and publish content identifiers using rpc_name_publish.
What's the difference between psa_add_pin and rpc_pin_add in Filebase (Web3 Storage) MCP? +
PSA is the high-level, dedicated Pinning Service API. RPC is the raw protocol call; use PSA unless you need to interact with the underlying IPFS daemon directly.
How can I check my total storage limit using Filebase (Web3 Storage) MCP? +
Run platform_get_usage. This gives a single metric that aggregates both your overall storage and bandwidth usage across all connected resources, so you don't have to guess where the limits are.
How do I manage multiple entry points or check my connection status using `platform_list_gateways`? +
You list all dedicated gateways with platform_list_gateways. This command provides an overview of every gateway you've set up, allowing you to see their current details and usage metrics. If you find a gateway that needs changes, you can then use platform_update_gateway or remove it entirely using platform_delete_gateway.
What is the difference between fetching raw data blocks with `rpc_block_get` and reading file contents using `rpc_cat`? +
The key difference is the output format. Use rpc_cat when you need to read a human-readable text file by its CID. If you use rpc_block_get, you're accessing the raw, low-level data blocks of the IPFS network; this is for debugging or advanced technical analysis.
I created an IPNS name pointing to old content; how do I update it without recreating the entire record using `platform_update_name`? +
You use platform_update_name to redirect your existing IPNS name. This is crucial because it lets you point that established name to a brand new CID, effectively updating its location without needing to delete and re-publish the resource entirely.
If I no longer use an old IPNS name or gateway, how do I safely remove these resources using `platform_delete_name` or `platform_delete_gateway`? +
You must explicitly delete abandoned resources. Use platform_delete_name to reclaim an unused IPNS entry, and run platform_delete_gateway when you're finished with a dedicated connection endpoint. This ensures proper cleanup of your decentralized infrastructure.
How do I pin a specific CID to ensure it stays on the network? +
You can use the rpc_pin_add tool by providing the CID. This ensures the content is persistently stored on Filebase's IPFS infrastructure.
Can I check my current storage usage and limits? +
Yes! Use the platform_get_usage tool. It will return your total storage used, bandwidth metrics, and current subscription limits.
How do I publish a CID to a mutable IPNS name? +
Use the rpc_name_publish tool with the target CID. This maps your content to an IPNS address that stays the same even when the content updates.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.