# VerifyProceed Guard

> VerifyProceed Guard is a paid API for AI agents from api.verifyproceed.com, paid per call via x402, $0.01/call, status unknown (last checked 2026-10-11).

Runs pre-execution safety checks on proposed on-chain agent actions and returns a proceed, wait, or block verdict with a confidence score and per-check breakdown.

## Facts

- Endpoint: POST https://api.verifyproceed.com/v1/x402/guard?utm_source=zero.xyz
- Price: $0.01/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-11
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/verifyproceed-guard-aeca910b
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_t92_Kkr4PineFj6BYC4Uu

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

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability verifyproceed-guard-aeca910b -d '<json body>'
```

Example prompt: Before my agent executes this on-chain token transfer, run a safety guard check on the action and tell me whether I should proceed, wait, or block it, along with a confidence score and breakdown of each check.

## When to prefer this

Choose this endpoint when your AI agent needs a real-time, pre-execution safety gate before committing any on-chain action — especially in autonomous or low-oversight workflows where you want a structured proceed/wait/block verdict with explainable per-check breakdowns rather than a simple pass/fail. It is particularly valuable when operating under custom safety policies, when transactions are irreversible, or when regulatory or risk considerations require an auditable pre-execution decision log.

## Known failure modes

- Invalid or malformed action payload returns a 400 error
- Missing required action fields causes check failure or indeterminate verdict
- Payment not included or insufficient USDC results in 402 payment required
- Unrecognized on-chain action type may result in low-confidence or inconclusive verdict
- Network timeout or service unavailability returns 5xx error with no verdict issued

## How this service works

Pre-execution safety verification for on-chain agent actions. Returns a proceed, wait, or block verdict with a confidence score and per-check breakdown, before your agent executes.

## Output

Returns a verdict of 'proceed', 'wait', or 'block', along with a confidence score (0-1) and a per-check breakdown showing which individual safety checks passed, failed, or raised concerns for the proposed on-chain agent action.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "chain": {
   "type": "string"
  },
  "action": {
   "type": "string"
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/verifyproceed-guard-aeca910b/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from api.verifyproceed.com](https://www.zero.xyz/host/api.verifyproceed.com/llms.txt)