Governance Playground
Click a scenario below and watch the governance chain respond in real time. No code, no API keys, no setup. See exactly how Curate-Me protects your AI agents.
Choose a scenario
Select a scenario above to see the governance chain in action
Every request passes through 6 policy checks in under 5ms
How the Governance Chain Works
Six policy checks run on every request. The chain short-circuits on the first denial -- no wasted compute, no data leaks.
Rate Limit
Enforces per-organization, per-key request limits. Returns IETF RateLimit-* headers so clients can self-throttle.
Cost Estimate
Calculates estimated cost from model + token count. Compares against per-request limits and daily budgets.
PII Scan
Regex-based scanner for SSNs, credit cards, API keys, emails, and other sensitive patterns. Blocks before data leaves your network.
Security Scanner
Detects prompt injection, jailbreak attempts, and data exfiltration patterns. No ML dependencies -- pure pattern matching.
Model Allowlist
Enforces which AI models each team can use. Prevents shadow AI and controls costs by model tier.
Human Approval
Flags high-cost or sensitive requests for human approval. Configurable thresholds per org.
Ready to Govern Your AI Agents?
Swap one URL, get all six guardrails. Zero code changes. Free tier includes 1,000 requests per day.