mcpField
A
AI Act Radar

AI Act Radar

Remote
aiactradar.comLegal & Government0Updated 12w ago

About AI Act Radar

AI Act Radar examples is a collection of runnable webhook receivers for processing AI Act Radar events across several platforms. It is for developers integrating webhook delivery; the examples share signature verification and idempotency handling.

Key features

  1. 01Receivers for six platforms
  2. 02HMAC-SHA-256 signature verification
  3. 03Timestamp freshness checks
  4. 04Idempotency key handling
  5. 05Signed sample-event test scripts

Use cases

  1. 01Integrating AI Act Radar webhooks into a platform
  2. 02Testing signed webhook events locally
  3. 03Building receivers that safely handle retries

MCP support

The README describes webhook receiver examples, not an MCP server or MCP integration.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://mcp.aiactradar.com/mcp/v1
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

All endpoints
Streamable HTTPhttps://mcp.aiactradar.com/mcp/v1Auth required

Tools

4
ParameterTypeRequiredDescription
limitintegerNo
sincestringNoISO 8601 timestamp; only events published after this. Optional.
untilstringNoISO 8601 timestamp; only events published before this. Optional.
topicsarrayNoTopic taxonomy. e.g. ["gpai","transparency","high-risk"]. Optional.
source_idstringNoOne of: eurlex, aioffice, aiboard, codeofpractice, haveyoursay, cen, bnetza, bfdi, bsi, cnil, nl-algoritmeregister, oecdai. Optional.
severity_minenumNoMinimum severity. Optional.info | low | medium | high | critical

Details

From the project README

aiactradar-examples

Reference webhook receivers for AI Act Radar. Each subdirectory is a self-contained, runnable example for one platform.

What every receiver does

  1. Reads the raw request body (signature is over the bytes, not the parsed JSON).
  2. Verifies X-AIAR-Timestamp is within 300 seconds of now.
  3. Verifies X-AIAR-Signature (v1=<hex>) using HMAC-SHA-256 over <timestamp>.<raw_body> in constant time.
  4. Checks an idempotency key (event.id) before doing real work — retries are normal.
  5. Returns 2xx within 5 seconds; otherwise the dispatcher retries with exponential backoff.

The verification is identical across platforms; only the platform glue differs.

Examples

FolderPlatformNotes
cloudflare-workers/Cloudflare WorkersWeb Crypto API, KV for idempotency
vercel-edge/Vercel Edge FunctionsWeb Crypto API, Vercel KV for idempotency
aws-lambda/AWS Lambda + API GatewayNode crypto module, DynamoDB for idempotency
express-node/Plain Express on NodeNode crypto, in-process LRU (replace for prod)
fastify-typescript/Fastify + TSTyped body parser, raw-body capture
deno-deploy/Deno DeployDeno KV for idempotency

Try it without signing up

Each example has a test.sh that signs a sample event with a local secret and posts it to the running receiver. You can run the full loop on your laptop in under five minutes:

cd cloudflare-workers
cp .dev.vars.example .dev.vars   # set AIA_SECRET to anything
npm install && npm run dev       # starts http://localhost:8787
# in another terminal:
bash test.sh                     # posts a signed sample event

Going to production

  • Store AIA_SECRET in your platform's secret manager. Never commit it.
  • Rotate the secret in the AI Act Radar dashboard at least every 90 days.
  • Subscribe to the dispatcher's webhook health channel — bounced deliveries surface there before customers notice.
  • Idempotency stores need a TTL of at least 7 days; the dispatcher retries up to 24h.

License

MIT.


AI Act Radar is operated by ligea GmbH, Karlsruhe. Information only — not legal advice.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.