How to Use the Hugging Face MCP in Google ADK

Q: How do I connect the Hugging Face MCP Server to Google ADK?

Run pip install google-adk to get the framework. Wrap your endpoint in McpToolset(serverparams=StreamableHttpServerParameters(url="...")) and pass it to the tools list in your LlmAgent setup.

Q: Can Google ADK restrict which Hugging Face tools my agent sees?

Yes. You pass an optional toolnames list when configuring the toolset. This lets you expose listmodels while hiding runtextgeneration from your production agent.

Q: Does this setup work with Gemini's long context window?

Yes. You can dump massive responses from getspace or listmodelsbyauthor straight into the context. Gemini processes the raw JSON and extracts exactly the parameters you need.

Q: How do I verify my Hugging Face API key in Google ADK?

Call the getaccount tool. If your environment variables are configured correctly, the server returns your profile details. A failed call means your agent lacks the necessary credentials.

Q: Is it safe to send corporate data through this MCP Server?

Tools like runinference push raw text prompts to external servers. Vinkius secures the transport layer with ephemeral, zero-trust execution, but you should never send proprietary codebase snippets to public endpoints.

Give your Google ADK agents direct access to Hugging Face datasets and remote inference APIs for enterprise workloads.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect Hugging Face MCP to Google ADK

Create your Vinkius account to connect Hugging Face to Google ADK 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 Hugging Face with Google ADK

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Map Hugging Face datasets to Google ADK

This MCP Server connects the Hugging Face hub directly to your Gemini-powered agent. Using `list_datasets` and `get_dataset`, your system searches for external data schemas to enrich internal BigQuery tables. The agent pulls metadata, reads the splits, and plans data ingestion routines. Long-context reasoning makes this extremely effective. Gemini can hold the entire JSON response from `list_collections` in memory, comparing hundreds of open-source datasets against your existing Vertex AI pipelines. It evaluates the structure before writing a single line of SQL.

Trigger specialized model inference

Generalist models handle reasoning, but `run_text_classification` and `run_summarization` offload specific tasks to specialized models. Your agent delegates narrow text processing jobs to external endpoints instead of consuming Gemini tokens. This keeps your Google Cloud compute costs down while maintaining high accuracy. You control exactly which endpoints the agent hits. By applying a `tool_names` filter to the `McpToolset` configuration, you restrict access to just `run_inference` or block it entirely. The agent only sees the operations you explicitly allow.

Audit open-source models dynamically

Finding the right model architecture requires the `list_models_by_task` and `get_model` tools. The agent queries the hub for specific capabilities, checking license types and download counts. It builds a manifest of viable candidates for your enterprise deployment. Verifying the network connection is handled by `check_hf_status`. Before the agent commits to a long-running data analysis job, it pings the API. If the hub goes down, the agent catches the error early and falls back to a different strategy.

Setup guide

Set up Hugging Face MCP in Google ADK

Prerequisites

Python 3.10+ installed
google-adk package (pip install google-adk)
Active Vinkius subscription with a valid endpoint token

1

Install Google ADK

Run pip install google-adk to install the Agent Development Kit. MCP support is included via the McpToolset class.
2

Connect via SSE transport

Use McpToolset.from_server() with SseServerParams pointing to your Vinkius endpoint. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com.
3

Create an LlmAgent

Pass the returned mcp_tools list directly to LlmAgent(tools=mcp_tools). The ADK maps each MCP tool to a native Gemini function call — no manual schema definitions required.
4

Run with any Gemini model

The agent works with any Gemini model (gemini-2.0-flash, gemini-2.5-pro, etc.). Copy the full example on the right to get started with Hugging Face tools in your ADK agent.

agent.py

from google.adk.agents import LlmAgent
from google.adk.tools.mcp_tool.mcp_toolset import McpToolset
from google.adk.tools.mcp_tool.mcp_session_manager import SseServerParams

# Connect to the MCP via SSE
mcp_tools, exit_stack = await McpToolset.from_server(
    connection_params=SseServerParams(
        url="https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
    )
)

# Create your agent with auto-discovered tools
agent = LlmAgent(
    name="Hugging Face_agent",
    model="gemini-2.0-flash",
    instruction="You have access to Hugging Face tools via MCP.",
    tools=mcp_tools,
)

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 Hugging Face. 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 Hugging Face 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 Hugging Face MCP in Google ADK

Run `pip install google-adk` to get the framework. Wrap your endpoint in `McpToolset(server_params=StreamableHttpServerParameters(url="..."))` and pass it to the `tools` list in your `LlmAgent` setup.

Yes. You pass an optional `tool_names` list when configuring the toolset. This lets you expose `list_models` while hiding `run_text_generation` from your production agent.

Yes. You can dump massive responses from `get_space` or `list_models_by_author` straight into the context. Gemini processes the raw JSON and extracts exactly the parameters you need.

Call the `get_account` tool. If your environment variables are configured correctly, the server returns your profile details. A failed call means your agent lacks the necessary credentials.

Tools like `run_inference` push raw text prompts to external servers. Vinkius secures the transport layer with ephemeral, zero-trust execution, but you should never send proprietary codebase snippets to public endpoints.

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