Databid

databid MCP server

Agent-native company intelligence. AI agents buy structured, verified company context on demand via API.

Live at: databid.io

What is this?

Companies list their operational context (capabilities, certifications, capacity, standards). AI agents query and buy that context via API or MCP. Think "SEO for the agent era."

Example: A procurement agent needs to evaluate a parts manufacturer → calls the API → gets structured data on materials, certifications, lead times, capacity → pays a few dollars → decision made in seconds.

Features

• MCP Server — Any MCP-compatible agent (Claude, Cursor, etc.) can connect at /api/mcp/mcp
• Agent Registration — Agents self-register, get API keys, submit requests
• Demand Signals — Platform tracks what agents are looking for, guiding companies on what to list
• Structured Context Packs — Standardized, token-efficient company data sized for agent context windows
• Verification — Platform verifies company claims (planned)

Tech Stack

• Next.js 16 (App Router) + TypeScript + Tailwind CSS v4
• PostgreSQL via Prisma ORM v7
• NextAuth.js (credentials + Google OAuth)
• MCP Server via mcp-handler
• Neon Postgres (production, via Vercel Marketplace)

Getting Started

Prerequisites

• Node.js 20+
• Docker (for PostgreSQL)

Setup

git clone https://github.com/indiano881/databid.git
cd databid
npm install

# Start PostgreSQL
docker run -d --name databid-postgres \
  -e POSTGRES_USER=postgres \
  -e POSTGRES_PASSWORD=postgres \
  -e POSTGRES_DB=databid \
  -p 5432:5432 postgres:16-alpine

# Set up environment variables
cp .env.example .env  # then edit .env with your values

# Run database migrations and start
npx prisma migrate dev
npm run dev

Open http://localhost:3000

Agent API

# Register and get an API key
curl -X POST https://databid.io/api/agents/register \
  -H "Content-Type: application/json" \
  -d '{"agentName":"MyBot","email":"[email protected]"}'

# Request a dataset
curl -X POST https://databid.io/api/requests \
  -H "Authorization: Bearer <your-api-key>" \
  -H "Content-Type: application/json" \
  -d '{"category":"manufacturing","description":"Steel suppliers in EU","keywords":["steel","EU"]}'

# Browse demand (public)
curl https://databid.io/api/requests/demand

MCP Server

Connect any MCP-compatible agent to https://databid.io/api/mcp/mcp (Streamable HTTP transport).

Claude Desktop — add to claude_desktop_config.json:

{
  "mcpServers": {
    "context-on-call": {
      "url": "https://databid.io/api/mcp/mcp"
    }
  }
}

Cursor — add to MCP settings:

{
  "mcpServers": {
    "context-on-call": {
      "url": "https://databid.io/api/mcp/mcp"
    }
  }
}

Available tools:

Tool	Description
register_agent	Register and get an API key for usage tracking
search_companies	Search company profiles by industry, location, certifications, capabilities
get_company_context	Get full structured context for a company
browse_demand	See what company context AI agents are requesting most
list_requests	List context requests with filters

Environment Variables

Variable	Required	Description
DATABASE_URL	Yes	PostgreSQL connection string
NEXTAUTH_URL	Yes	App URL (http://localhost:3000 for dev)
NEXTAUTH_SECRET	Yes	Random secret for JWT signing
GOOGLE_CLIENT_ID	No	Google OAuth client ID
GOOGLE_CLIENT_SECRET	No	Google OAuth client secret

License

MIT