MCP Face Transform Server

A Model Context Protocol (MCP) server that exposes the AceDataCloud Face Transform API — face keypoint detection, beautification, age/gender transform, face swap, cartoonization, and liveness detection.

Status: All Face APIs are currently in Alpha. Interfaces may evolve.

Features

• Keypoint detection — 90+ landmarks per face, multi-face supported
• Beautification — smoothing, whitening, face slimming, eye enlarging
• Age transform — age or de-age a portrait
• Gender transform — swap perceived facial gender characteristics
• Face swap — move a source face onto a target image (with optional async webhook)
• Cartoonize — render a portrait in animated / cartoon style
• Liveness detection — distinguish live captures from printed / screen photos

Connect: hosted OAuth, API token, or local stdio

The hosted endpoint is https://face.mcp.acedata.cloud/mcp. Choose one route for the MCP client:

The hosted service advertises OAuth metadata and Dynamic Client Registration (DCR). DCR registers the client application; it is not an API key. OAuth signs you in and the client sends the resulting Bearer token; it may reuse or create an API credential for the account. Browser sign-in still requires an AceDataCloud account. The hosted service can be metered: review current service documentation and displayed pricing before a real operation. Do not configure both an OAuth login and a fixed Authorization header for the same server.

Hosted OAuth examples

• Claude and Claude Desktop chat: Add a remote custom connector in Customize → Connectors → Add custom connector, enter https://face.mcp.acedata.cloud/mcp, select sign-in, and choose Register automatically if Claude asks how to register its OAuth client. Complete consent. Claude Desktop's local claude_desktop_config.json is a separate setup. Claude connector guide.
• Claude Code: claude mcp add --transport http --scope user face https://face.mcp.acedata.cloud/mcp, then claude mcp login face. Check /mcp. Claude Code MCP guide.
• Cursor: Add a remote server with only https://face.mcp.acedata.cloud/mcp. For a project, merge the entry below into <project>/.cursor/mcp.json; for personal use, use ~/.cursor/mcp.json. Cursor MCP guide.
• VS Code / Copilot: Run MCP: Add Server, select HTTP, enter https://face.mcp.acedata.cloud/mcp, then finish the browser sign-in. New portable workspace configs use <project>/.mcp.json; the VS Code-specific format below uses <project>/.vscode/mcp.json or the user profile. Check MCP: List Servers. VS Code MCP setup.
• Codex: codex mcp add face --url https://face.mcp.acedata.cloud/mcp, then codex mcp login face. Its user settings are in ~/.codex/config.toml. Official Codex MCP guide.

Cursor project config (OAuth):

{
  "mcpServers": {
    "face": {"url": "https://face.mcp.acedata.cloud/mcp"}
  }
}

VS Code-specific workspace config (OAuth):

{
  "servers": {
    "face": {"type": "http", "url": "https://face.mcp.acedata.cloud/mcp"}
  }
}

Hosted API token

Sign in at AceDataCloud Platform, open the service page, and obtain an API credential. A fixed Bearer header is useful when your client lacks OAuth; an invalid header does not fall back to OAuth in Claude Code. The header value is sensitive, so keep it out of committed files and screenshots.

For Claude Code, the shell expands the token when you add the server; treat the saved user MCP config as a secret:

export ACEDATACLOUD_API_TOKEN='YOUR_API_TOKEN'
claude mcp add --transport http --scope user face https://face.mcp.acedata.cloud/mcp \
  --header "Authorization: Bearer $ACEDATACLOUD_API_TOKEN"

For a Claude Code project config, put a variable reference in <project>/.mcp.json and set that variable in the environment that launches Claude Code:

{
  "mcpServers": {
    "face": {
      "type": "http",
      "url": "https://face.mcp.acedata.cloud/mcp",
      "headers": {"Authorization": "Bearer ${ACEDATACLOUD_API_TOKEN}"}
    }
  }
}

Cursor uses a different environment-variable syntax in ~/.cursor/mcp.json or an uncommitted project config:

{
  "mcpServers": {
    "face": {
      "url": "https://face.mcp.acedata.cloud/mcp",
      "headers": {"Authorization": "Bearer ${env:ACEDATACLOUD_API_TOKEN}"}
    }
  }
}

In VS Code, run MCP: Open User Configuration and merge this server plus its masked input; ${input:...} is for VS Code's user/workspace format and is not portable to the Agent Host .mcp.json format:

{
  "inputs": [
    {"id": "acedata-face-token", "type": "promptString", "description": "AceDataCloud API token", "password": true}
  ],
  "servers": {
    "face": {
      "type": "http",
      "url": "https://face.mcp.acedata.cloud/mcp",
      "headers": {"Authorization": "Bearer ${input:acedata-face-token}"}
    }
  }
}

For Cline, use its MCP configuration UI or CLI file ~/.cline/data/settings/cline_mcp_settings.json; its remote transport value is streamableHttp. For JetBrains AI Assistant, add a remote URL from Settings → Tools → AI Assistant → Model Context Protocol (MCP). For Zed, use a context_servers entry with the URL only for OAuth or add a local Bearer header. These clients have different configuration schemas; follow their current UI rather than copying another client's JSON. Cline · JetBrains · Zed.

Local stdio

Install the package and give the local process an API token:

python -m pip install mcp-face-transform
export ACEDATACLOUD_API_TOKEN='YOUR_API_TOKEN'
mcp-face-transform

For Claude Desktop local MCP, merge this entry into the file opened by its developer settings (~/Library/Application Support/Claude/claude_desktop_config.json on macOS). uvx requires uv on PATH:

{
  "mcpServers": {
    "face": {
      "command": "uvx",
      "args": ["mcp-face-transform"],
      "env": {"ACEDATACLOUD_API_TOKEN": "YOUR_API_TOKEN"}
    }
  }
}

Keep this user-level file private. Self-hosted HTTP uses mcp-face-transform --transport http --port 8000; expose it only with suitable network and TLS controls. Local execution still calls the AceDataCloud API.

Check before using the service

1. https://face.mcp.acedata.cloud/health returning {"status":"ok"} checks endpoint reachability only.
2. Confirm that the MCP client loads tools. face_get_usage_guide is a reference tool; it does not verify downstream API access or balance.
3. Use a documented service operation for an end-to-end check after reviewing current service documentation.

For 401, check which auth route the client used and whether the token or OAuth session is valid. A 403 may mean an account permission or content moderation failure; read the returned error. Insufficient balance and downstream service failures need their own diagnosis. A listed tool or submitted task does not prove a successful result.

Tool Reference

Example

"Detect all faces in https://example.com/group.jpg and return their keypoints."
→ face_detect_keypoints(image_url="https://example.com/group.jpg")

"Lighten and smooth my portrait."
→ face_beautify(image_url="https://example.com/me.jpg", smoothing=15, whitening=25)

"Replace the face in the scene with the headshot."
→ face_swap(
    source_image_url="https://example.com/headshot.jpg",
    target_image_url="https://example.com/scene.jpg",
  )

Configuration in Claude Desktop / Claude Code

{
  "mcpServers": {
    "face-transform": {
      "command": "uvx",
      "args": ["mcp-face-transform"],
      "env": {
        "ACEDATACLOUD_API_TOKEN": "your_api_token_here"
      }
    }
  }
}

Or use the hosted endpoint with bearer auth:

{
  "mcpServers": {
    "face-transform": {
      "url": "https://face.mcp.acedata.cloud/mcp",
      "headers": {
        "Authorization": "Bearer your_api_token_here"
      }
    }
  }
}

Development

pip install -e ".[dev,test]"
pytest --cov=core --cov=tools
ruff check .

Service details

Service details

License

MIT — see LICENSE.