AI LAW RADAR · Daily Last verified 26 Jun 2026

Developers & agents

Build on the live register

The same primary-sourced AI-law data behind the Radar, available two ways: a free read-only JSON API for code, and an MCP endpoint so your AI agents can query it live. Both return facts and official source URLs — never legal conclusions. Open data under CC BY 4.0; just credit AI Law Radar.

REST API

/api/v1 · JSON · CORS-enabled

Plain GET requests, JSON responses, no key required for reasonable read use. Every record carries its effective_date, status, max_penalty, confidence, a checked date and an official source_url.

EndpointReturnsQuery params
GET /api/v1/obligationsFilter the registerregion, role, topic, status, q
GET /api/v1/deadlinesUpcoming dated deadlines, soonest firstregion, role, topic, within
GET /api/v1/jurisdictionsTracked regions + flagship laws
GET /api/v1/changesDated change logsince
GET /api/v1/metaCounts + last-verified date
# Obligations that bind an EU deployer
curl "https://ailawradar.com/api/v1/obligations?region=EU&role=deployer"

# Upcoming deadlines in the next 90 days
curl "https://ailawradar.com/api/v1/deadlines?within=90"

# Every tracked jurisdiction + its flagship law
curl "https://ailawradar.com/api/v1/jurisdictions"

# Changes since a date
curl "https://ailawradar.com/api/v1/changes?since=2026-06-01"

MCP — for your AI agents

streamable HTTP

AI Law Radar speaks the Model Context Protocol, so assistants like Claude can pull live AI-law facts mid-conversation. Point any MCP client at the endpoint below:

https://ailawradar.com/api/mcp

In a config that takes remote MCP servers (e.g. Claude Desktop, Cursor):

{
  "mcpServers": {
    "ai-law-radar": {
      "type": "http",
      "url": "https://ailawradar.com/api/mcp"
    }
  }
}

For a client that only speaks stdio, bridge it: npx mcp-remote https://ailawradar.com/api/mcp. Tools exposed:

Or just take the data

static · CC BY 4.0

No API needed — the whole register is published as static files: obligations.json, jurisdictions.json, changelog.json, meta.json. Change feeds: RSS · JSON. Machine summary: llms.txt.