Vaquill Integrations

Chatbots, widgets, extensions, MCP servers, and platform connectors for Vaquill AI — the legal research API.

Each integration is self-contained — pick any one, deploy it independently.

Quick Start

All chatbots and widgets use the Vaquill API. You need a Vaquill API key (vq_key_...) from vaquill.ai/dashboard.

git clone https://github.com/Vaquill-AI/integrations.git
cd integrations/<integration-name>
cp .env.example .env   # add your VAQUILL_API_KEY
# follow the integration's README for setup

Chatbots

All chatbots share the same architecture:

• Vaquill API client (vaquill_client.py) for legal Q&A
• Per-user/channel conversation history (multi-turn)
• Rate limiting (Redis or in-memory)
• Structured source citations from case law
• Slash commands, feedback buttons, analytics

Widgets

Drop-in chat widgets for any website. Embed with a script tag or iframe.

Browser Extension

Chrome extension popup with legal AI chat. Calls the Vaquill API directly.

Automation

Pre-built workflow templates for batch legal research via spreadsheets.

MCP Servers

Hosted MCP endpoints for AI tool integration:

Claude Desktop / Claude Code

{
  "mcpServers": {
    "courtlistener": {
      "url": "https://courtlistener-mcp.vaquill.ai/mcp/"
    },
    "canlii": {
      "url": "https://canlii-mcp.vaquill.ai/mcp",
      "headers": { "Authorization": "Bearer YOUR_TOKEN" }
    }
  }
}

courtlistener-mcp

US federal and state court opinions, dockets, PACER data, and eCFR federal regulations via the CourtListener API v4.

• Language: Python (FastMCP)
• Transport: streamable_http at /mcp/ (default), sse, stdio
• Docker: docker-compose up -d (port 8000)
• API key: Free from courtlistener.com/api

canlii-mcp

Canadian federal and provincial court decisions and legislation via the CanLII API.

• Language: TypeScript (MCP SDK + Hono)
• Transport: streamable_http at /mcp (stateless), stdio
• Docker: docker run -e CANLII_API=key -p 3000:3000 (port 3000)
• API key: Apply at canlii.org/en/api

Vaquill API

All integrations use the same simple API:

curl -X POST https://api.vaquill.ai/api/v1/ask \
  -H "Authorization: Bearer vq_key_..." \
  -H "Content-Type: application/json" \
  -d '{"question": "What is Section 302 IPC?"}'

Response:

{
  "data": {
    "answer": "Section 302 of the Indian Penal Code deals with...",
    "sources": [
      {
        "sourceIndex": 1,
        "caseName": "State of UP v. Ram Sagar Yadav",
        "citation": "(1985) 1 SCC 552",
        "court": "Supreme Court of India",
        "excerpt": "...",
        "relevanceScore": 0.94
      }
    ],
    "mode": "standard"
  },
  "meta": {
    "processingTimeMs": 2340.5,
    "creditsConsumed": 0.5,
    "creditsRemaining": 4.5
  }
}

See the API Reference for full documentation.

License

MIT. Each MCP server retains its upstream license — see individual subdirectories.