# Services from guardroute.ai on Zero

> guardroute.ai serves 6 paid API services for AI agents, indexed by Zero. Each is callable per call with payment handled by Zero.

Canonical page: https://www.zero.xyz/host/guardroute.ai
Live health (JSON, refreshed every minute): https://www.zero.xyz/host/guardroute.ai/health.json

## Services

- [GuardRoute.ai Step-Up Approval](https://www.zero.xyz/c/guardroute-ai-step-up-approval-dc7a4601/llms.txt): $0.025/call via x402. Performs live risk scoring and policy routing on a pending action (wire, beneficiary change, workflow step, or AI-agent request) and returns an ALLOW or escalate decision before execution.
- [GuardRoute.ai Route Simulation](https://www.zero.xyz/c/guardroute-ai-route-simulation-f61ce8ef/llms.txt): $0.025/call via x402. Simulates pre-execution risk scoring and policy routing decisions for financial transactions, beneficiary changes, and AI-agent actions before they are executed.
- [GuardRoute.ai Budget Enforcer](https://www.zero.xyz/c/guardroute-ai-budget-enforcer-5d1a733f/llms.txt): $0.025/call via x402. Performs live risk scoring and policy routing on AI-agent requests, enforcing budget and spend limits before any action is executed.
- [GuardRoute.ai Tool Call Policy Check](https://www.zero.xyz/c/guardroute-ai-tool-call-policy-check-84ead6eb/llms.txt): $0.025/call via x402. Returns a live risk score and policy routing decision (ALLOW or block) for an AI-agent tool call before it is executed.
- [GuardRoute.ai Fail-Closed Gate](https://www.zero.xyz/c/guardroute-ai-fail-closed-gate-e5712dd8/llms.txt): $0.025/call via x402. Evaluates an incoming action or request against risk policies and returns a deterministic routing decision (allow/deny) before the action is executed.
- [GuardRoute.ai Risk Scoring & Policy Routing](https://www.zero.xyz/c/guardroute-ai-risk-scoring-policy-routing-d0f25338/llms.txt): $0.025/call via x402. Performs live risk scoring and policy routing for financial transactions, beneficiary changes, workflow actions, and AI-agent requests before they are executed.

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

- [Zero catalog index](https://www.zero.xyz/llms.txt)