Lucairn is an EU-based privacy gateway that sits between your MCP-enabled agent and the upstream LLM (Claude, GPT-4o, o1/o3/o4).
What it does:
- Automatically detects and replaces PII — names, emails, IBANs, addresses, phone numbers, customer IDs — with placeholders before reaching the upstream LLM. The model never sees the raw values.
- Each response carries a signed compliance certificate (ed25519 + RFC 3161 timestamp + Sigstore Rekor inclusion proof) that proves what was redacted, when, and by which sanitizer layer.
- Cross-provider BYOK: bring your own ANTHROPIC_API_KEY and/or OPENAI_API_KEY. Routing is automatic by model prefix (claude-* → Anthropic, gpt-* / o1-* → OpenAI).
- Free Developer tier with 1,000 requests/month. Pro and Enterprise tiers unlock higher quotas and re-linkage of placeholders to original values in the response.
Exposes one MCP tool, `chat_via_lucairn`, using the Anthropic Messages API wire format. Compatible with any MCP client (Claude Desktop, Cline, Continue, Cursor, etc.).
Source: https://github.com/Declade/lucairn-sdks
Glama listing: https://glama.ai/mcp/servers/Declade/lucairn-sdks
Docs: https://lucairn.eu
概要
Lucairn Privacy Gateway — MCP Server
EU-based privacy gateway for AI agents. Lucairn sits between your MCP-enabled client and upstream OpenAI or Anthropic models, pseudonymizes PII before the model sees it, and returns a signed verification receipt for each request.
Free Developer tier: 500 requests/month. Bring your own provider key with ANTHROPIC_API_KEY, OPENAI_API_KEY, or both. No credit card required.
Quick Start
npx -y @lucairn/mcp-server
The server speaks MCP over stdio and exposes one tool, chat_via_lucairn.
Configure Your MCP Client
Add this to your MCP config file, such as claude_desktop_config.json, cline_mcp_settings.json, or a Cursor/Continue MCP config:
{
"mcpServers": {
"lucairn": {
"command": "npx",
"args": ["-y", "@lucairn/mcp-server@latest"],
"env": {
"LUCAIRN_API_KEY": "<your_lucairn_api_key>",
"ANTHROPIC_API_KEY": "<optional_byok_anthropic_key>",
"OPENAI_API_KEY": "<optional_byok_openai_key>"
}
}
}
}
Restart your MCP client. The chat_via_lucairn tool should appear automatically.
Get a Lucairn API Key
Create a free Developer account:
https://lucairn.eu/account/signup
Developer is BYOK and includes 500 requests/month through the hosted gateway.
What the Server Does
Your agent calls chat_via_lucairn with a messages-style payload: model, messages, max_tokens, and optional system / temperature.
Lucairn scans user content for personal data such as names, emails, IBANs, addresses, phone numbers, customer IDs, and similar identifiers.
Detected PII is replaced with placeholders such as [PERSON_1], [EMAIL_2], or [IBAN_3] before the request reaches the upstream model.
The upstream model receives only the sanitized prompt.
Lucairn returns the model response plus a signed verification receipt showing what was detected and pseudonymized.
Developer-tier responses keep placeholders visible. Pro and Enterprise can support relinking workflows.
Provider Routing
Lucairn routes from the model parameter and the BYOK keys you provide.
Model prefix Routes to Env var used
claude-*, anthropic-* Anthropic ANTHROPIC_API_KEY
gpt-*, openai-*, o1-*, o3-*, o4-* OpenAI OPENAI_API_KEY
Use any supported OpenAI or Anthropic model available to your provider account. If you only set one provider key, call models from that provider.
Links
Source: https://github.com/Declade/lucairn-sdks
npm: https://www.npmjs.com/package/@lucairn/mcp-server
Setup docs: https://lucairn.eu/developer/mcp
Sign up: https://lucairn.eu/account/signup
Verify receipts: https://lucairn.eu/verify
Glama listing: https://glama.ai/mcp/servers/Declade/lucairn-sdks
License
MIT — see https://github.com/Declade/lucairn-sdks/blob/main/LICENSE
サーバー設定
{
"mcpServers": {
"lucairn": {
"url": "https://gateway.lucairn.eu/mcp",
"headers": {
"Authorization": "Bearer <YOUR_LUCAIRN_API_KEY>"
}
}
}
}