How to Use the Fireblocks MCP in Pydantic AI

Q: How do I initialize the Fireblocks MCP Server toolset within my Pydantic AI code?

You use MCPToolset("http://...") pointing to your Vinkius MCP endpoint, then pass it to the Agent constructor in the toolsets parameter for automatic tool discovery.

Q: How do I estimate transaction fees safely with this framework?

Your agent calls estimatefee or estimatenetworkfee to get structured fee estimates, which are parsed into typed models for accurate gas limit calculations.

Run type-safe digital asset workflows in Pydantic AI using this Fireblocks MCP Server integration to eliminate runtime errors.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect Fireblocks MCP to Pydantic AI

Create your Vinkius account to connect Fireblocks 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 Fireblocks with Pydantic AI

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Type-Safe Vault Operations

Creating secure storage begins with `create_vault_account` which your Pydantic AI agent validates against strict python schemas before execution. The tool returns structured data that the framework parses immediately, ensuring that any missing fields in the vault response trigger a validation error rather than a silent failure. You then use `create_vault_account_asset` to initialize specific tokens with guaranteed type safety. If you need to generate deposit addresses, the agent runs `create_vault_account_asset_address`. Every address payload is parsed by Pydantic models, protecting your downstream databases from malformed cryptographic strings.

Validated Transaction Lifecycle in Pydantic AI

Initiating transfers requires `create_transaction` which your Pydantic AI agent executes using strictly typed transaction parameters. The framework guarantees that gas limits, destination addresses, and asset IDs match the exact Fireblocks API schema before the request leaves your server. This prevents costly on-chain failures caused by malformed arguments. To check the transfer, the agent queries `get_transaction` and maps the response to a Pydantic model. If the status changes, you receive a typed object, allowing your python logic to react safely without parsing raw JSON strings.

Whitelist and Contract Verification

Adding smart contracts to your whitelist requires `add_contract` which your Pydantic AI agent executes after checking the contract address format. The agent queries `list_contracts` to inspect existing records and ensures no duplicate entries exist in your local state. This runtime type checking prevents your agent from interacting with unverified or malformed contract addresses. When interacting with verified dApps, the agent calls `get_contract_asset` to confirm asset support. Because Pydantic AI enforces strict data models, you catch API changes instantly during runtime.

Setup guide

Set up Fireblocks 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": {
        "fireblocks-mcp": {
            "url": "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
        }
    }
})

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

result = await agent.run("List recent Fireblocks 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 Fireblocks. 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 Fireblocks 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 Fireblocks MCP in Pydantic AI

The framework validates all inputs against strict Pydantic models before executing tools via the MCP Server. If your agent attempts to pass an invalid address, Pydantic AI raises a validation error immediately.

If the Fireblocks API returns a payload that deviates from the schema, Pydantic AI fails loudly with a validation error, preventing corrupt data from entering your treasury workflow.

You use `MCPToolset("http://...")` pointing to your Vinkius MCP endpoint, then pass it to the `Agent` constructor in the `toolsets` parameter for automatic tool discovery.

Your agent calls `estimate_fee` or `estimate_network_fee` to get structured fee estimates, which are parsed into typed models for accurate gas limit calculations.

Your API keys and transaction histories are never stored or tracked. The server runs in an ephemeral, zero-trust V8 isolate that processes your requests in memory and discards the state immediately after completion.

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