Quickstart

Get RADAR evaluating in under 5 minutes. Deploy the container, connect evidence collection, and verify your first evidence record.

PULL AND RUN

Pull the RADAR Docker image and start evaluating immediately. No license key required for evaluation — the free self-hosted version includes full evidence collection.

bash

# Pull and start RADAR
docker run -d --name radar -p 8080:8080 \
  akioudai/radar:latest

# Verify it is running
curl http://localhost:8080/health

Auto-generated trial

No license file required — a 30-day trial license is auto-generated on first run. Includes core compliance (EU AI Act, GDPR, SOC 2) with full evidence collection.

CONNECT EVIDENCE

Point RADAR at your LLM gateway or agent telemetry. Use the CLI to configure a source.

bash

# Configure your LLM gateway as an evidence source
docker exec radar radar sources add --type gateway \
  --name my-agents \
  --endpoint http://your-gateway:4000

# Verify connection
docker exec radar radar status
# Expected: Sources: 1 connected | Traces: 0 | Status: active

RADAR supports gateway log ingestion, webhook receivers, REST API push, OpenTelemetry, and log file parsing. See integration guides for your stack.

VIEW EVIDENCE

Once RADAR detects agent activity, open the dashboard or use the CLI to inspect traces and findings.

bash

# Open the dashboard
open http://localhost:8080

# Or check via CLI
docker exec radar radar traces list --limit 5
docker exec radar radar findings list --severity high

EXPORT A PACK

Generate a regulator-ready evidence pack from collected traces.

bash

# Export evidence as a compliance pack
docker exec radar radar pack --last 24h \
  --format html --output /tmp/evidence-pack.html

# Open the report
open /tmp/evidence-pack.html

Next Steps#

Full Evaluation Guide

Step-by-step deployment, source configuration, and evidence review

Evidence Model

Understand traces, findings, control mappings, and export bundles

FAQ

Comparisons, evaluation limits, deployment questions