How to Use the Backblaze B2 MCP in Pydantic AI

Q: What happens if my Pydantic AI agent gets an unexpected API response from Backblaze B2?

It will raise a ValidationError. Instead of letting your agent work with corrupted or incomplete data, Pydantic AI stops execution and tells you exactly how the response from the Backblaze B2 tool failed validation.

Build brutally reliable agents for Backblaze B2 with Pydantic AI, where every API response is validated against a strict schema.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect Backblaze B2 MCP to Pydantic AI

Create your Vinkius account to connect Backblaze B2 to Pydantic AI and route execution through our secure gateway. The platform manages server hosting, runtime updates, and security layers. Configuration requires no manual server provisioning.

GDPR Free for Subscribers

Setup Backblaze B2 with Pydantic AI

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Type-Safe Bucket Operations

Stop guessing what the API will return. When your Pydantic AI agent calls `list_buckets`, the JSON response from the MCP Server is parsed and validated against a Pydantic model at runtime. You're guaranteed to get a list of proper bucket objects, or you get a loud, immediate `ValidationError`. This means you can write agent logic that trusts its inputs. Use `create_bucket` to provision new storage and `get_file_info` to check an object's checksum, knowing that malformed data will never silently corrupt your agent's state. If the B2 API changes unexpectedly, your agent will fail safely instead of running with bad data.

Reliable File & Version Control

Manage files with certainty. The `hide_file` tool lets your agent perform a soft delete, and Pydantic AI ensures the response confirms the action was successful. If the API call fails for any reason, your agent knows, because the error response won't match the expected success model. This is critical for destructive actions like `delete_file_version`. You can build an automated cleanup agent that removes old file versions and have confidence that it's working correctly. The type-safe structure prevents your agent from assuming a file was deleted when it wasn't.

Build Failsafe Deletion Logic with This MCP Server

This is where correctness really matters. The `delete_bucket` tool is designed to fail if the target bucket contains any files. A generic agent might just see a 400 Bad Request and not know why. But Pydantic AI can be configured to expect this specific error structure. Your agent doesn't just fail; it fails intelligently. It knows the deletion was rejected because the bucket wasn't empty, and can then decide to either list and delete the files first or report the issue. It turns a simple API error into actionable intelligence for your agent.

Setup guide

Set up Backblaze B2 MCP in Pydantic AI

Prerequisites

Python 3.10+ installed
pydantic-ai-slim[fastmcp] package
Active Vinkius subscription with a valid endpoint token

1

Install Pydantic AI with FastMCP
Run pip install "pydantic-ai-slim[fastmcp]". The FastMCP toolset replaces the deprecated MCPServerHTTP class with full protocol support.
2

Configure the FastMCPToolset
Pass a JSON-style config dict to FastMCPToolset with your Vinkius URL. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com. Supports Streamable HTTP, SSE, and Stdio transports.
3

Create and run your agent
Pass the toolset to Agent(toolsets=[toolset]) and call agent.run(). Swap openai:gpt-4o for any supported model — Anthropic, Google, Mistral, or Groq.

agent.py

from pydantic_ai import Agent
from pydantic_ai.toolsets.fastmcp import FastMCPToolset

toolset = FastMCPToolset({
    "mcpServers": {
        "backblaze-b2-mcp": {
            "url": "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
        }
    }
})

agent = Agent(
    "openai:gpt-4o",
    toolsets=[toolset],
    system_prompt="You have access to Backblaze B2 tools.",
)

result = await agent.run("List recent Backblaze B2 transactions")
print(result.output)

Get your connection token →

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.

Why Choose Vinkius

Vinkius connects your tools to AI with real-time monitoring and automatic cost savings — all from one dashboard.

Connect Backblaze B2 now

Real-time monitoring

Live

visibility into every interaction

Connect your favorite tools to your AI and see exactly what's happening — every request, every response, in real time.

Built-in savings

60%

lower AI costs

Vinkius compresses data between your apps and your AI automatically. Lower bills every month — no configuration required.

Single dashboard

One

place for every integration

Every tool your AI connects to, managed from a single screen. One account, complete control.

Common questions about Backblaze B2 MCP in Pydantic AI

Pydantic AI validates the raw JSON response from the MCP Server against a predefined Pydantic model. If the response doesn't perfectly match the expected schema—for example, if a field is missing or has the wrong type—it raises a `ValidationError` immediately.

It will raise a `ValidationError`. Instead of letting your agent work with corrupted or incomplete data, Pydantic AI stops execution and tells you exactly how the response from the Backblaze B2 tool failed validation.

Yes, that's a perfect use case. Your Pydantic AI agent can call the `delete_file_version` tool, and the framework's validation ensures the agent gets a clear, correct confirmation that the operation succeeded on your Backblaze B2 account.

No. Pydantic AI is model-agnostic. You can use it with models from OpenAI, Anthropic, Google, or even a local model you're running yourself.

It's limited to your account's metadata: bucket names, file lists, and object details like checksums and headers. The Vinkius server that runs the tools operates in a zero-trust, ephemeral sandbox. It never sees your actual file content or raw credentials.

Use it with your favorite AI tools

Connect this server to Cursor, Claude, VS Code, and more.

OpenAI Agents SDK sdk-python

Google ADK sdk-python

Pydantic AI sdk-python

Vercel AI SDK sdk-typescript