Backblaze B2 MCP. Manage object storage and access rules directly in your chat.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
Backblaze B2 MCP Server. Manage your cloud storage architecture through natural conversation. Use this server to control buckets, file versions, and objects in Backblaze B2.
Your AI agent can list buckets, delete file versions, check for failed uploads, and audit object metadata without you ever leaving your editor.
What your AI agents can do
Authorize account
Finds the specific regional API endpoint assigned to your billing account.
Create bucket
Provisions a new logical storage bucket partition.
Delete bucket
Permanently deletes an empty Backblaze Storage Bucket.
List all existing buckets, create new ones, or change the access rules (public/private) on any bucket.
Soft-delete files by marking them as hidden, or permanently remove specific file versions to reclaim storage space.
Scan the B2 nodes to find incomplete or stalled multipart uploads, helping diagnose broken pipelines.
Retrieve detailed information for any file, including its SHA1 hash, content type, and size.
Get a paginated list of object names that are currently visible and not hidden by lifecycle rules.
Ask AI about this MCP
Supported MCP Clients
Gemini Waiting for input…
Backblaze B2 MCP Server: 10 Tools for Cloud Storage
These 10 tools let your AI agent perform full CRUD operations on your Backblaze B2 buckets, objects, and file versions.
019d7557authorize account
Finds the specific regional API endpoint assigned to your billing account.
019d7557create bucket
Provisions a new logical storage bucket partition.
019d7557delete bucket
Permanently deletes an empty Backblaze Storage Bucket.
019d7557delete file version
Removes a specific version of a file from disk storage.
019d7557get file info
Gets the content type, size, and SHA1 hash for a specific file.
019d7557hide file
Marks an active B2 file as hidden, effectively performing a soft delete.
019d7557list buckets
Retrieves a list of all existing storage buckets on your account.
019d7557list file names
Gets a list of primary object metadata from a specific bucket.
019d7557list unfinished large files
Scans B2 nodes for incomplete multipart upload chunks.
019d7557update bucket
Changes the global access control list (ACL) privacy settings for a bucket.
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 Backblaze B2, then connect any of our 4,700+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 4,700+ 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
What you can do with this MCP connector
You'll control your Backblaze B2 storage right from your AI client. You can manage your whole cloud setup using natural conversation.
Audit and control storage buckets: You can list all existing buckets (list_buckets), create a new storage bucket partition (create_bucket), and delete empty buckets (delete_bucket). You'll also update a bucket's access rules, changing its global access control list privacy settings (update_bucket).
Manage file versions and deletions: You can soft-delete files by marking them as hidden (hide_file), or permanently remove specific file versions to reclaim space (delete_file_version).
Inspect object metadata: You can get detailed information for any file, checking its content type, size, and SHA1 hash (get_file_info).
List accessible file names: You can retrieve a list of object names visible within a specific bucket (list_file_names).
Check for upload failures: You can scan B2 nodes to find incomplete or stalled multipart uploads, which helps you diagnose broken pipelines (list_unfinished_large_files).
Authentication: You'll first run authorize_account to find the specific regional API endpoint assigned to your billing account.
How Backblaze B2 MCP Works
- 1 Subscribe to the server and input your Backblaze B2 Application Key ID and Key.
- 2 Ask your AI agent a direct question about your cloud storage (e.g., 'What buckets do we have?').
- 3 The agent calls the appropriate tool, and the server returns the data or confirmation, completing the action.
The bottom line is: your agent uses the server to execute complex cloud storage commands without needing you to write any API code.
Who Is Backblaze B2 MCP For?
This is for the DevOps engineer who needs to audit storage state across dozens of buckets before a deployment. It's for the Backend Developer needing to provision a private asset domain instantly. It's for the System Admin who needs to check object integrity across thousands of files without manual scripting.
Quickly verify bucket states, clean up abandoned file versions, and debug failed multipart streams after a major deployment.
Create and secure private storage domains or retrieve file hashes to validate assets before deploying an application component.
Audit the object footprint, check file integrity, and manage access rules for large, multi-tenant storage environments.
What Changes When You Connect
- Instantly audit all storage containers. Use
list_bucketsto see every bucket name and its current privacy status without running a separate CLI command. - Fix data clutter with targeted deletions. Use
delete_file_versionto remove specific, old versions of a file, preventing storage bloat. - Debug failed uploads immediately. Run
list_unfinished_large_filesto check for stalled multipart chunks that are eating up space. - Validate asset integrity. Run
get_file_infoto pull the SHA1 hash and content type of any object, ensuring the file hasn't been tampered with. - Secure your infrastructure. Use
update_bucketto change a bucket's ACL from public to private with a simple prompt.
Real-World Use Cases
Checking for abandoned data chunks
The ops engineer noticed file uploads failing intermittently. Instead of digging through API logs, they prompt: 'Check for unfinished large files.' The agent runs list_unfinished_large_files and reports the exact buckets and chunks that need manual attention, solving the debugging headache.
Creating a new secure log repository
A backend developer needs a dedicated, private bucket for new logs. They ask the agent to 'Create a new private bucket named X.' The agent uses create_bucket and sets the access rule to allPrivate instantly, bypassing manual console setup.
Archiving old, public assets
A system admin finds a bucket full of outdated public assets. They ask the agent to 'Hide all files in this bucket.' The agent uses hide_file, which soft-deletes the objects, making them invisible to standard listing commands while keeping the data for a later sweep.
Verifying object ownership before migration
A team needs to move a set of assets. They ask the agent to 'Get the file details and SHA1 for index.html.' The agent runs get_file_info, providing the required SHA1 hash and size, which is critical for the migration manifest.
The Tradeoffs
Sequential API calls
Trying to check a bucket's status, then listing files, and then updating permissions by running three different commands manually. This takes minutes and fails if the first command times out.
→
Ask your agent to perform the entire sequence: 'List the buckets, and for the one named 'assets', make it private.' The agent orchestrates list_buckets, then update_bucket in a single, continuous session.
Forgetting to check for versions
Simply deleting a file via the console assumes the latest version is the only one that matters, potentially leaving old, sensitive versions accessible and costing money.
→
Use list_file_names to see the active files, then specifically use delete_file_version to target and remove the exact version ID you want gone.
Ignoring API endpoints
Assuming every bucket uses the same API endpoint, leading to connection failures when the storage provider changes the regional endpoint.
→
Always start by running authorize_account first. This ensures your agent uses the correct, dynamically assigned API endpoint for your billing region.
When It Fits, When It Doesn't
Use this server if your job involves managing state across multiple storage buckets (e.g., dev, staging, prod) and you need to audit, create, or modify access controls. It's ideal for DevOps and System Admins. Don't use it if your only task is to upload a single file—use a dedicated upload tool instead. If you only need to check if a file exists, list_file_names is sufficient. If you need to change the file's security settings, use update_bucket. If you're worried about data loss, always run list_file_names before running any delete command.
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Backblaze B2. 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 10 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Available Capabilities
Manual Cloud Auditing is a Click-Heavy Nightmare
Right now, checking a single bucket's status means jumping between the console's listing view, the version history tab, and the access control policy panel. You have to click 'View Details,' then copy the ID, and then paste it into a separate audit script. It's slow, tedious, and easy to miss a version or a hidden object.
With the Backblaze B2 MCP Server, you just ask your agent: 'What's wrong with the assets bucket?' It runs the necessary checks—listing files, checking versions, and flagging incomplete uploads—and gives you the full audit report in one response. You stop clicking and start solving.
Backblaze B2 MCP Server: Manage object storage and access rules
You no longer have to write multi-step scripts to clean up old data. Instead of logging in, running a delete command, then checking the API logs for failure codes, you tell your agent: 'Remove all versions of the old logs in the 'archive' bucket.'
The agent handles the whole workflow. It executes `delete_file_version` and confirms the removal, making the entire process atomic and conversational. It's direct. It's fast.
Common Questions About Backblaze B2 MCP
How do I check if my bucket is public or private using the Backblaze B2 MCP Server? +
You can list the buckets and their access rules using list_buckets. This tool returns the current privacy status (e.g., allPublic or allPrivate) for every storage partition.
What is the difference between `hide_file` and `delete_file_version` with Backblaze B2 MCP Server? +
hide_file performs a soft delete, marking the file as invisible but retaining the data for future recovery or sweeps. delete_file_version is a hard delete, which permanently removes the specific file chunk from the disk array.
Can I find broken file uploads using the Backblaze B2 MCP Server? +
Yes. You run list_unfinished_large_files. This tool scans the B2 nodes and reports any stalled multipart upload chunks, letting you fix broken pipelines.
Does the Backblaze B2 MCP Server handle access rules for my buckets? +
Yes. You use update_bucket to mutate the global ACL privacy settings, allowing you to change a bucket from public to private with a simple command.
How do I use `authorize_account` to validate my Backblaze B2 API endpoint? +
The authorize_account tool finds your dynamic API URL (e.g., api003.backblazeb2.com). This confirms which regional endpoint your billing account uses for all operations.
What is the purpose of `list_file_names` when I only want to see active files? +
list_file_names retrieves only the primary, surface-level object metadata. This skips hidden files or versions that might be filtered out by standard listing commands.
Which tool should I use to make a bucket completely empty before deletion? +
You must ensure all lifecycle rules have purged content before running delete_bucket. The tool fails with a 400 error if any files or hidden versions remain.
Can `get_file_info` provide enough data to check file integrity? +
Yes, get_file_info returns the SHA1 checksum and content type for a specific file. This data lets you validate the object's integrity against a known hash.
Can my AI agent debug broken uploads in Backblaze B2? +
Yes. Ask your agent to list unfinished large files. It detects multipart upload shards that disconnected before completion, which consume billing quota. It provides granular visibility so you can safely clear dead streams via API.
Can the agent make my private bucket publicly accessible to serve CDN assets? +
Yes. If commanded, the agent can update the bucket privacy setting from 'allPrivate' to 'allPublic'. This instantly routes objects to be fetched securely using friendly URLs for front-end hosting or CDN origins.
How does the agent handle file deletion safely using B2 protocol? +
The agent can perform both soft deletes (hide file marker, triggering lifecycle purge later) or absolute hard version deletes. If you tell the agent to 'hide the file', the file remains physically safe but vanishes from directory listings.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.
More in this category
New Relic
Monitor and query your entire stack via New Relic NerdGraph — track entities, NRQL, and alerts directly from your AI agent.
Buildkite
Automate CI/CD pipelines via Buildkite — manage builds, agents, and deployments directly from any AI agent.
Sirv
Enable your AI agent to manage files, read metadata, and monitor bandwidth usage on your Sirv CDN account.
You might also like
Harry Potter API
Explore the wizarding world — search characters, students, staff, houses and spells from the Harry Potter universe.
Amilia
Recreation and activity management — manage programs, accounts, and registrations via AI.
MOBIDI
Manage your mobile advertising campaigns with real-time bidding, audience targeting, and performance analytics for app installs.