# CAPI2 Agent Commerce – Action Guard

> CAPI2 Agent Commerce – Action Guard is a paid API for AI agents from capi2-agent-commerce.vandurmedries.workers.dev, paid per call via x402, $0.02/call, status unknown (last checked 2026-09-15).

Performs programmable authority checks on proposed autonomous agent actions and returns tamper-evident, risk-scored decisions with detailed findings.

## Facts

- Endpoint: POST https://capi2-agent-commerce.vandurmedries.workers.dev/v1/action-guard
- Price: $0.02/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-15
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/capi2-agent-commerce-action-guard-79bed94a
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_C61PS01jHSU5CLANRUK23

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 capi2-agent-commerce-action-guard-79bed94a -d '<json body>'
```

Example prompt: Before my agent executes this purchase action, run an authority check on it — pass the authority object and the proposed action, and tell me the decision, any findings like price_mismatch, and the risk score so I know whether to proceed.

## When to prefer this

Choose this endpoint when you need a programmatic, per-action authorization gate for autonomous agents — especially when you require tamper-evident decision receipts and structured risk scores rather than a simple boolean allow/deny. It is particularly suited for agentic pipelines where agents are performing financial or high-stakes actions and you want fine-grained findings (e.g. price_mismatch codes) rather than coarse policy enforcement.

## Known failure modes

- Malformed authority or proposed_action objects may result in validation errors or unexpected block decisions
- High risk scores may be returned for actions that are actually legitimate if authority rules are misconfigured
- x402 payment failure at $0.02 USDC will prevent the check from executing
- Ambiguous or incomplete action schemas may lead to inconclusive findings
- Service availability depends on Cloudflare Workers uptime

## How this service works

Programmable authority checks and tamper-evident receipts for autonomous agents, sold through sustainable x402 commerce.

## Output

A JSON object containing a 'decision' field ('allow' or 'block'), a 'findings' array of objects each with a 'code' indicating the specific policy violation (e.g. 'price_mismatch'), and a 'risk_score' integer (0–100) reflecting the overall danger level of the proposed action.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "authority": {
   "type": "object"
  },
  "proposed_action": {
   "type": "object"
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "decision": "block",
  "findings": [
   {
    "code": "price_mismatch"
   }
  ],
  "risk_score": 100
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/capi2-agent-commerce-action-guard-79bed94a/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from capi2-agent-commerce.vandurmedries.workers.dev](https://www.zero.xyz/host/capi2-agent-commerce.vandurmedries.workers.dev/llms.txt)