# Hugging Face MCP connector

The Hugging Face connector brings model, dataset and Space repositories into Elaichi, so Claude, ChatGPT, Cursor and the Elaichi Agent can search the Hub, read model cards and browse repository files as the signed-in person.

Source: https://elaichi.ai/connectors/huggingface/

## Facts

| | |
| --- | --- |
| Application | Hugging Face |
| Category | Artificial Intelligence |
| AI tools | 4 |
| Authentication | Connects over OAuth |
| Bring your own OAuth app | No |
| Native MCP | Yes. Hugging Face builds and runs this MCP server. Elaichi adds sign-in, access controls and an audit log on top |
| Support for its tools | huggingface.co/support |
| MCP endpoint | https://api.elaichi.ai/mcp |
| Works with | Claude, ChatGPT, Cursor, any MCP client, and the Elaichi Agent |
| Tools advertised by name | No. Connected tools are never listed one by one, however few there are. The endpoint advertises `search_tools` and `execute_tool` instead |

## What you can ask once Hugging Face is connected

- Find Apache licensed text classification models with the most downloads
- Which French sentiment datasets have more than 10k downloads
- List the files in our organization's latest model repository

## Connect Hugging Face in Elaichi

This happens once for the organization, before any client is involved.

1. Open Connections, choose Add connection, and pick Hugging Face.
2. Optionally set Share with, then press Connect.
3. Approve it in Hugging Face. Hugging Face's own window opens. Whoever approves it decides what this connection can reach.

Credentials are vaulted and nobody, including the AI, reads them back. The connection becomes a toolbox immediately, so you can curate which Hugging Face tools are exposed, rename them, or freeze arguments before anyone points a client at it.

## Hugging Face MCP connector for Claude

Endpoint: https://api.elaichi.ai/mcp

1. Open Customize, then Connectors.
2. Press Add.
3. Name it, paste the MCP server URL, then Continue.
4. Sign in and approve.

On Team and Enterprise, an Owner adds it once. Everyone else turns it on for themselves.

## Hugging Face MCP connector for ChatGPT

Endpoint: https://api.elaichi.ai/mcp

1. Open Plugins, then press the + button.
2. Name it and paste the endpoint into Server URL.
3. Leave Authentication on OAuth, then tick the risk acknowledgement.
4. Press Create, then sign in and approve.

Works on the web today. The plugin directory lives at chatgpt.com/plugins.

## Hugging Face MCP connector for Cursor

Endpoint: https://api.elaichi.ai/mcp

1. Open `~/.cursor/mcp.json`.
2. Add the endpoint under `mcpServers`.
3. Reload Cursor, then sign in and approve.

Set up per machine, so repeat it on each computer you work from.

## Connect Hugging Face to any MCP client

Endpoint: https://api.elaichi.ai/mcp

1. Add the endpoint as a remote MCP server.
2. Sign in and approve.

The Elaichi Agent already has these tools, with nothing to set up.

## What the consent screen decides

Only Read is granted by default, which is not enough to call a Hugging Face tool. Over MCP there is no trusted place to confirm a write in the moment, so the consent screen is the standing approval rather than a formality. Grant Read and Run tools. Think hard before granting Delete, which reaches into connected apps and cannot be undone.

## What teams do with Hugging Face through Elaichi

### Shortlist candidate models before a build

ML engineering. Ask for the most downloaded text classification models on Hugging Face under an Apache license, then compare their model cards without opening a dozen tabs.

### Find a dataset that fits the task

Data science. Search the Hub for datasets by task, language and size, then read the dataset card and look through the files before you commit to anything.

### Read a model card without leaving the chat

Research. Pull the details of a specific repository, including its tags, license, downloads and recent updates, and have the agent summarize what matters for your paper or experiment.

### Check a model's license before anyone ships it

Compliance. Have the agent confirm the license on each Hugging Face repository your teams are using and flag any that need a closer look.

### Explain what a model does in plain language

Product. Ask for a non-technical summary of a model's card so a product review can move forward without waiting on an engineer.

### Keep track of your organization's public repositories

Developer relations. List the models, datasets and Spaces published under your organization and read their cards to see which ones are out of date.

## Elaichi vs Zapier MCP vs Composio for Hugging Face

All three can connect Hugging Face to an AI assistant, and all three have admin controls. They differ in where access lives and how you pay.

| What to check | Elaichi | Zapier MCP | Composio |
| --- | --- | --- | --- |
| Where the AI connects | One address for the whole organization. Endpoint: https://api.elaichi.ai/mcp | A server per member, created at sign-in. | An MCP endpoint per team, or an SDK. |
| Control over Hugging Face tools | Allow or restrict single Hugging Face tools, per role or user. | App and action restrictions on the account. | Role permissions, down to the action. |
| Record of calls | One audit entry per Hugging Face call. | A History tab of tool calls. | A log of every tool call. |
| Single sign-on | SAML or OIDC, plus SCIM, on Gold. | SAML on Enterprise. | SAML and OIDC on Enterprise. |
| Price | $15 per user per month. | 2 tasks per successful call. | Billed per tool call. |

Sources: Zapier MCP [docs](https://docs.zapier.com/mcp/get-started/quickstart), [security](https://docs.zapier.com/mcp/manage/security), [usage](https://docs.zapier.com/mcp/features/usage); Composio [docs](https://docs.composio.dev/docs/composio-connect), [gateway](https://composio.dev/mcp-gateway), [enterprise](https://composio.dev/enterprise), [pricing](https://composio.dev/pricing). Checked September 2026.

Longer take: [Zapier MCP alternative](/blog/zapier-mcp-alternative/) and [when you don't need an MCP gateway](/blog/when-you-dont-need-an-mcp-gateway/).

## Frequently asked questions

### How do I connect Hugging Face to Claude?

Connect Hugging Face in Elaichi first: pick it from the catalog and sign in with your Hugging Face account over OAuth, approving the access it asks for. There is no OAuth application to register and no client ID or secret to generate. Then in Claude open Customize, then Connectors, then Add, and paste https://api.elaichi.ai/mcp. Claude signs you in through Elaichi and Hugging Face is available in your next conversation.

### Does Hugging Face work with ChatGPT and Cursor as well as Claude?

Yes. Once Hugging Face is connected in Elaichi, the same endpoint, https://api.elaichi.ai/mcp, works in Claude, ChatGPT, Cursor, any other MCP client and the Elaichi Agent. You connect Hugging Face once and every client you use picks it up.

### What can an AI agent actually do with my Hugging Face data?

With Hugging Face connected, an agent can search the Hub for models, datasets and Spaces, read a repository's card, license, tags and download counts, browse the files inside a repository, and confirm which Hugging Face account it is working as. Short, concrete asks work best, such as the license of a named model or the most downloaded datasets for a task, rather than long multi-part sentences.

### Does connecting Hugging Face give the AI access to all my repositories and organizations?

No. Every call to Hugging Face runs as the person who signed in, so the agent sees only the public repositories and the private repositories and organizations that person can already open on Hugging Face. Elaichi can narrow that further with roles and restrictions, but it can never widen what Hugging Face itself allows.

### Can my team share one Hugging Face connection?

Yes. One person connects Hugging Face in Elaichi and shares the connection with a team, and nobody else ever handles a token. Each teammate still signs in to Elaichi as themselves, so the audit log names the person who searched the Hub or read a repository, not a shared account.

### Can I stop an agent from changing things in Hugging Face?

Yes. Restrictions in Elaichi apply per action on the Hugging Face connector, so you can leave searching and reading open while closing off anything you do not want an agent to touch. A restricted action is never advertised to Claude, ChatGPT or Cursor at all, so no prompt can reach it.

### What happens to a Hugging Face connection when someone leaves?

Offboarding a person in Elaichi ends their access to Hugging Face through every client at once. A Hugging Face connection shared with a team keeps working for everyone else, and if you disconnect Hugging Face in Elaichi it disappears from Claude, ChatGPT, Cursor and the Elaichi Agent in one step.

### Does the Hugging Face MCP connector work with Gemini, Codex, Claude Code or other MCP clients?

Yes. Hugging Face is reached over the same MCP endpoint every client uses, so anything that speaks MCP can call it — Gemini, Codex, Claude Code, Windsurf, Cline, Zed and OpenCode among them — alongside Claude, ChatGPT, Cursor, and the Elaichi Agent. The tools on offer and the access behind them are identical whichever client asks. Only the setup screen differs.

### Is Elaichi an alternative to Zapier MCP for Hugging Face?

Yes. Both let Claude, ChatGPT or Cursor use Hugging Face. Zapier MCP fits a team that already automates in Zapier, since each person signs in and acts as themselves in that account. Elaichi fits when IT wants one address for the whole company, per-tool rules by role, and a record of every Hugging Face call.

### How is Elaichi different from Composio for Hugging Face?

Composio gives AI agents tools and sign-in handling across 1,000+ apps, for developers building agents or people using an assistant, billed per tool call. Elaichi gives a company's own people governed access to Hugging Face: one address, restrictions per role or user, and $15 per user per month. Both have role permissions and a log of every call.

## All 4 Hugging Face tools

Every tool below is callable through https://api.elaichi.ai/mcp once Hugging Face is connected, subject to the toolbox it is in and the restrictions on the caller.

- **Hf whoami** (Action). Inspect the current Hugging Face authentication context, including the account, visible organization memberships, and credential access details. Read-only and never returns credential values.
- **Hub repo search** (Search). Search Hugging Face repositories with a shared query interface. You can target models, datasets, spaces, or aggregate across multiple repo types in one call. Include links to repositories in your response.
- **Hub repo details** (Action). Get details for one or more Hugging Face repos (model, dataset, or space). Auto-detects type unless specified. For datasets, use operations: overview, dataset_structure, dataset_preview. Use dataset_structure first to discover configs, splits, sizes, and schema. Use…
- **Hf fs** (Action). When to use: Hugging Face Hub models, datasets, Spaces, collections, papers, daily papers, today's trending models, current paper leaderboard, docs, and repository files. Examples: {"operations":[{"cmd":"ls","args":["hf://models/trending","--limit","10"]}]}…