MCP server (experimental)
Connect Claude Code, Cursor, VS Code and other agents to Quba over the Model Context Protocol.
Experimental
The MCP server is experimental. Tools and their schemas may change before a stable release.
Quba runs a remote MCP server at https://www.quba.co/api/mcp. Agents can scan and anonymize text, read your organization's policies and check the balance. Every call uses an API key and is billed like the API.
Tools
| Tool | What it does |
|---|---|
scan_text | Find personal data in text, with each match's type, position and confidence. |
anonymize_text | Remove or transform personal data, with inline rules or the policy assigned to your key. |
list_policies | List your organization's policies. |
get_policy | Read a policy's rules. |
get_balance | Credits left for your organization. |
Connect
Claude Code
claude mcp add --transport http quba https://www.quba.co/api/mcp \
--header "Authorization: Bearer <your-api-key>"Cursor
Add to ~/.cursor/mcp.json:
{
"mcpServers": {
"quba": {
"url": "https://www.quba.co/api/mcp",
"headers": { "Authorization": "Bearer <your-api-key>" }
}
}
}VS Code
Add to .vscode/mcp.json:
{
"inputs": [
{ "type": "promptString", "id": "quba-key", "description": "Quba API key", "password": true }
],
"servers": {
"quba": {
"type": "http",
"url": "https://www.quba.co/api/mcp",
"headers": { "Authorization": "Bearer ${input:quba-key}" }
}
}
}Other clients
Any client that supports Streamable HTTP works: use the URL above and send the key as Authorization: Bearer <key> or x-api-key: <key>.
Good to know
- Text an agent sends to a tool is already part of the conversation. Use the tools to clean text before it leaves: sharing, logs, other tools.
- Quba processes the text and does not store it.
- Policies are read-only over MCP. Create and edit them in the dashboard.
- claude.ai and ChatGPT connectors are not supported yet.
Anonymization
Learn how to use Quba's Anonymization API to anonymize sensitive data in text using various techniques like replacement, masking, hashing, and encryption.
Scan Text POST
Scan text for sensitive data. Runs the configured detection entities over the request text and returns every match as a character range with its entity type and confidence score. Detections below the confidence threshold are omitted.