# Trifecta Security Score — AI Agent Risk Assessment

> Trifecta Security Score — AI Agent Risk Assessment is a paid API for AI agents from tollbooth.memerhuwhite.workers.dev, paid per call via x402, $0.05/call, status unknown (last checked 2026-11-16).

Computes a security risk trifecta score for an AI agent by evaluating whether it handles private data, performs outbound actions, and processes untrusted content, optionally accounting for compensating controls.

## Facts

- Endpoint: POST https://tollbooth.memerhuwhite.workers.dev/paid/security/trifecta-score?utm_source=zero.xyz
- Price: $0.05/call
- Payment: x402
- Status: unknown
- Last checked: 2026-11-16
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/trifecta-security-score-ai-agent-risk-assessment-2959b308
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_x6QTXoldYd0P5vNfE_7GR

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 trifecta-security-score-ai-agent-risk-assessment-2959b308 -d '<json body>'
```

Example prompt: Can you give me the trifecta security score for my agent — it does handle private user data, it makes outbound API calls, and it processes untrusted content from the web, but I've put rate limiting and sandboxing in place as compensating controls.

## When to prefer this

Use this endpoint when you need a quick, structured security risk rating for an AI agent or agentic workflow based on the three canonical risk dimensions: private data exposure, outbound action capability, and untrusted content ingestion. It is especially useful during agent design reviews, compliance checks, or automated CI/CD security gates for agentic systems. Prefer this over manual review when you need a consistent, comparable numeric score across multiple agent configurations.

## Known failure modes

- Missing required boolean fields (has_private_data, has_outbound_actions, has_untrusted_content) returns a validation error
- Payment failure via x402 protocol returns 402 Payment Required
- Invalid compensating_controls format (non-array) may cause a 400 error
- Service unavailability on the Cloudflare Worker returns 503

## How this service works

100+ paid x402 MCP tools for AI agents. Prediction markets, SEC filings, federal data, OSINT, academic research, health, environmental. Pay per call in USDC on Base.

## Output

Returns a structured object with a trifecta security score and metadata including source, generation timestamp, success status, and whether the result was cached. The score reflects the combination of risky capabilities (private data, outbound actions, untrusted content) modulated by any compensating controls provided.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "required": [
  "has_private_data",
  "has_outbound_actions",
  "has_untrusted_content"
 ],
 "properties": {
  "has_private_data": {
   "type": "boolean"
  },
  "has_outbound_actions": {
   "type": "boolean"
  },
  "compensating_controls": {
   "type": "array"
  },
  "has_untrusted_content": {
   "type": "boolean"
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "data": {
   "oneOf": [
    {
     "type": "array"
    },
    {
     "type": "object"
    }
   ]
  },
  "meta": {
   "type": "object",
   "properties": {
    "count": {
     "type": "integer"
    },
    "source": {
     "type": "string"
    },
    "generated_at": {
     "type": "string",
     "format": "date-time"
    }
   }
  },
  "cached": {
   "type": "boolean"
  },
  "success": {
   "type": "boolean"
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/trifecta-security-score-ai-agent-risk-assessment-2959b308/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from tollbooth.memerhuwhite.workers.dev](https://www.zero.xyz/host/tollbooth.memerhuwhite.workers.dev/llms.txt)
